End to End Learning for Self-Driving Cars

Abstract

We trained a convolutional neural network (CNN) to map raw pixels from a single front-facing camera directly to steering commands. This end-to-end approach proved surprisingly powerful. With minimum training data from humans the system learns to drive in traffic on local roads with or without lane markings and on highways. It also operates in areas with unclear visual guidance such as in parking lots and on unpaved roads. The system automatically learns internal representations of the necessary processing steps such as detecting useful road features with only the human steering angle as the training signal. We never explicitly trained it to detect, for example, the outline of roads. Compared to explicit decomposition of the problem, such as lane marking detection, path planning, and control, our end-to-end system optimizes all processing steps simultaneously. We argue that this will eventually lead to better performance and smaller systems. Better performance will result because the internal components self-optimize to maximize overall system performance, instead of optimizing human-selected intermediate criteria, e.g., lane detection. Such criteria understandably are selected for ease of human interpretation which doesn't automatically guarantee maximum system performance. Smaller networks are possible because the system learns to solve the problem with the minimal number of processing steps. We used an NVIDIA DevBox and Torch 7 for training and an NVIDIA DRIVE(TM) PX self-driving car computer also running Torch 7 for determining where to drive. The system operates at 30 frames per second (FPS).

Cited by in corpus (701)

• A Survey of Autonomous Driving: Common Practices and Emerging Technologies
• A Survey of Deep Learning Techniques for Autonomous Driving
• On Calibration of Modern Neural Networks
• Deep Multi-modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges
• An Introduction to Deep Reinforcement Learning
• DeepXplore: Automated Whitebox Testing of Deep Learning Systems
• Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
• A Survey on Distributed Machine Learning
• Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems
• SeqSleepNet: End-to-End Hierarchical Recurrent Neural Network for Sequence-to-Sequence Automatic Sleep Staging
• Motion Planning for Autonomous Driving: The State of the Art and Future Perspectives
• How to Train Your Robot with Deep Reinforcement Learning; Lessons We've Learned
• The Hidden Vulnerability of Distributed Learning in Byzantium
• Guiding Deep Learning System Testing using Surprise Adequacy
• Countering Adversarial Images using Input Transformations
• SciANN: A Keras/Tensorflow wrapper for scientific computations and physics-informed deep learning using artificial neural networks
• Deep Predictive Coding Networks for Video Prediction and Unsupervised Learning
• AMP: Adversarial Motion Priors for Stylized Physics-Based Character Control
• Safe, Multi-Agent, Reinforcement Learning for Autonomous Driving
• Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning
• Deep Neural Networks for Data-Driven Turbulence Models
• Explanations in Autonomous Driving: A Survey
• Multimodal End-to-End Autonomous Driving
• Adversarial Perturbations Against Deep Neural Networks for Malware Classification
• DLFuzz: Differential Fuzzing Testing of Deep Learning Systems
• Adversarial Attack and Defense on Graph Data: A Survey
• A Survey of End-to-End Driving: Architectures and Training Methods
• One-Shot Visual Imitation Learning via Meta-Learning
• A2D2: Audi Autonomous Driving Dataset
• Testing Scenario Library Generation for Connected and Automated Vehicles, Part I: Methodology
• Collision Avoidance in Pedestrian-Rich Environments with Deep Reinforcement Learning
• Explaining How a Deep Neural Network Trained with End-to-End Learning Steers a Car
• MIT Advanced Vehicle Technology Study: Large-Scale Naturalistic Driving Study of Driver Behavior and Interaction with Automation
• A Survey of Deep RL and IL for Autonomous Driving Policy Learning
• DeepTest: Automated Testing of Deep-Neural-Network-driven Autonomous Cars
• Demystifying Parallel and Distributed Deep Learning: An In-Depth Concurrency Analysis
• Flow: A Modular Learning Framework for Mixed Autonomy Traffic
• Multi-modal Sensor Fusion-Based Deep Neural Network for End-to-end Autonomous Driving with Scene Understanding
• Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift
• Pervasive AI for IoT applications: A Survey on Resource-efficient Distributed Artificial Intelligence
• Sim4CV: A Photo-Realistic Simulator for Computer Vision Applications
• DARTS: Deceiving Autonomous Cars with Toxic Signs
• Are Labels Required for Improving Adversarial Robustness?
• Testing Deep Neural Networks
• Bringing AI To Edge: From Deep Learning's Perspective
• Adversarial Machine Learning in Image Classification: A Survey Towards the Defender's Perspective
• Causal Confusion in Imitation Learning
• Virtual to Real Reinforcement Learning for Autonomous Driving
• Evaluating Bayesian Deep Learning Methods for Semantic Segmentation
• A nonlocal physics-informed deep learning framework using the peridynamic differential operator
• End-to-End Deep Reinforcement Learning for Lane Keeping Assist
• Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examples
• Enable Deep Learning on Mobile Devices: Methods, Systems, and Applications
• InfoGAIL: Interpretable Imitation Learning from Visual Demonstrations
• TensorFuzz: Debugging Neural Networks with Coverage-Guided Fuzzing
• TABOR: A Highly Accurate Approach to Inspecting and Restoring Trojan Backdoors in AI Systems
• Faster CNNs with Direct Sparse Convolutions and Guided Pruning
• Testing Scenario Library Generation for Connected and Automated Vehicles, Part II: Case Studies
• COOPERNAUT: End-to-End Driving with Cooperative Perception for Networked Vehicles
• Variational End-to-End Navigation and Localization
• Adversarial Examples, Uncertainty, and Transfer Testing Robustness in Gaussian Process Hybrid Deep Networks
• Towards Proving the Adversarial Robustness of Deep Neural Networks
• Image coding for machines: an end-to-end learned approach
• PyTorch Distributed: Experiences on Accelerating Data Parallel Training
• Query-Efficient Imitation Learning for End-to-End Autonomous Driving
• Data Science Methodologies: Current Challenges and Future Approaches
• LookOut: Diverse Multi-Future Prediction and Planning for Self-Driving
• A Game-Based Approximate Verification of Deep Neural Networks with Provable Guarantees
• Multi Sensor Fusion for Navigation and Mapping in Autonomous Vehicles: Accurate Localization in Urban Environments
• A Survey on Approximate Edge AI for Energy Efficient Autonomous Driving Services
• Hardware Acceleration of Sparse and Irregular Tensor Computations of ML Models: A Survey and Insights
• Backdoor Embedding in Convolutional Neural Network Models via Invisible Perturbation
• An Analysis of ISO 26262: Using Machine Learning Safely in Automotive Software
• StarNet: Targeted Computation for Object Detection in Point Clouds
• Mind the Gap! A Study on the Transferability of Virtual vs Physical-world Testing of Autonomous Driving Systems
• Provably Minimally-Distorted Adversarial Examples
• Grasp2Vec: Learning Object Representations from Self-Supervised Grasping
• Reinforcement Learning from Imperfect Demonstrations
• Verification for Machine Learning, Autonomy, and Neural Networks Survey
• Generative Adversarial Imitation from Observation
• Practical Blind Membership Inference Attack via Differential Comparisons
• F1/10: An Open-Source Autonomous Cyber-Physical Platform
• Controlling an Autonomous Vehicle with Deep Reinforcement Learning
• Optimization and Abstraction: A Synergistic Approach for Analyzing Neural Network Robustness
• Controlling Steering Angle for Cooperative Self-driving Vehicles utilizing CNN and LSTM-based Deep Networks
• Self-Driving Car Steering Angle Prediction Based on Image Recognition
• Artificial Neural Networks for Photonic Applications: From Algorithms to Implementation
• Model-Predictive Policy Learning with Uncertainty Regularization for Driving in Dense Traffic
• Language-guided Navigation via Cross-Modal Grounding and Alternate Adversarial Learning
• Evaluating Uncertainty Quantification in End-to-End Autonomous Driving Control
• Spatial Action Maps for Mobile Manipulation
• PRIMA: General and Precise Neural Network Certification via Scalable Convex Hull Approximations
• Intention-Net: Integrating Planning and Deep Learning for Goal-Directed Autonomous Navigation
• Learning to Drive using Inverse Reinforcement Learning and Deep Q-Networks
• Sensitivity Analysis of Deep Neural Networks
• A Survey of Deep Network Solutions for Learning Control in Robotics: From Reinforcement to Imitation
• End-to-End Deep Learning for Steering Autonomous Vehicles Considering Temporal Dependencies
• Towards Practical Verification of Machine Learning: The Case of Computer Vision Systems
• Reachable Set Computation and Safety Verification for Neural Networks with ReLU Activations
• Data-based stabilization of unknown bilinear systems with guaranteed basin of attraction
• FusionPlanner: A Multi-task Motion Planner for Mining Trucks via Multi-sensor Fusion
• Revisiting the Calibration of Modern Neural Networks
• Social Attention for Autonomous Decision-Making in Dense Traffic
• How much real data do we actually need: Analyzing object detection performance using synthetic and real data
• First Three Years of the International Verification of Neural Networks Competition (VNN-COMP)
• Recurrent Neural Networks for Semantic Instance Segmentation
• An LSTM-Based Autonomous Driving Model Using Waymo Open Dataset
• Scanner: Efficient Video Analysis at Scale
• DeepRoad: GAN-based Metamorphic Autonomous Driving System Testing
• Waymo Driverless Car Data Analysis and Driving Modeling using CNN and LSTM
• Combinatorial Testing for Deep Learning Systems
• Freeway Merging in Congested Traffic based on Multipolicy Decision Making with Passive Actor Critic
• Learned Visual Navigation for Under-Canopy Agricultural Robots
• Testing and verification of neural-network-based safety-critical control software: A systematic literature review
• Off-road Autonomous Vehicles Traversability Analysis and Trajectory Planning Based on Deep Inverse Reinforcement Learning
• TBD: Benchmarking and Analyzing Deep Neural Network Training
• SQIL: Imitation Learning via Reinforcement Learning with Sparse Rewards
• Automated Verification of Neural Networks: Advances, Challenges and Perspectives
• Deep Imitation Learning for Complex Manipulation Tasks from Virtual Reality Teleoperation
• Detecting Adversarial Examples by Input Transformations, Defense Perturbations, and Voting
• Learning End-to-end Multimodal Sensor Policies for Autonomous Navigation
• Falsification and consciousness
• MODNet: Moving Object Detection Network with Motion and Appearance for Autonomous Driving
• Hierarchical Program-Triggered Reinforcement Learning Agents For Automated Driving
• Causal Learning and Explanation of Deep Neural Networks via Autoencoded Activations
• Modified DDPG car-following model with a real-world human driving experience with CARLA simulator
• VisualBackProp: efficient visualization of CNNs
• SUNRISE: A Simple Unified Framework for Ensemble Learning in Deep Reinforcement Learning
• Investigating Deep Learning Methods for Obtaining Photometric Redshift Estimations from Images
• Benchmark and application of unsupervised classification approaches for univariate data
• Polarity Loss for Zero-shot Object Detection
• The MineRL 2019 Competition on Sample Efficient Reinforcement Learning using Human Priors
• Learning Non-Metric Visual Similarity for Image Retrieval
• Reachability Analysis and Safety Verification for Neural Network Control Systems
• DeepBillboard: Systematic Physical-World Testing of Autonomous Driving Systems
• iPlanner: Imperative Path Planning
• On Adversarial Examples and Stealth Attacks in Artificial Intelligence Systems
• A Survey on Scenario-Based Testing for Automated Driving Systems in High-Fidelity Simulation
• Control Strategies for Autonomous Vehicles
• End-to-end Multi-Modal Multi-Task Vehicle Control for Self-Driving Cars with Visual Perception
• Deep Q-Network Based Decision Making for Autonomous Driving
• Path Integral Networks: End-to-End Differentiable Optimal Control
• Estimation of Driver's Gaze Region from Head Position and Orientation using Probabilistic Confidence Regions
• Scalable End-to-End Autonomous Vehicle Testing via Rare-event Simulation
• Finite Versus Infinite Neural Networks: an Empirical Study
• End-to-end Learning of Driving Models from Large-scale Video Datasets
• Using Machine Learning Safely in Automotive Software: An Assessment and Adaption of Software Process Requirements in ISO 26262
• Blockchain-Powered Collaboration in Heterogeneous Swarms of Robots
• Adversarial Driving: Attacking End-to-End Autonomous Driving
• Emergence of Exploratory Look-Around Behaviors through Active Observation Completion
• ZhuSuan: A Library for Bayesian Deep Learning
• Uncertainty-aware Short-term Motion Prediction of Traffic Actors for Autonomous Driving
• Reinforcement Learning for Robotic Manipulation using Simulated Locomotion Demonstrations
• Visual Attention Prediction Improves Performance of Autonomous Drone Racing Agents
• DNNV: A Framework for Deep Neural Network Verification
• Robust Backdoor Attacks against Deep Neural Networks in Real Physical World
• Calibration of Deep Probabilistic Models with Decoupled Bayesian Neural Networks
• APPLE: Adaptive Planner Parameter Learning from Evaluative Feedback
• Imitation from Observation: Learning to Imitate Behaviors from Raw Video via Context Translation
• Detecting Adversarial Attacks on Neural Network Policies with Visual Foresight
• From WiscKey to Bourbon: A Learned Index for Log-Structured Merge Trees
• Systematic Testing of Convolutional Neural Networks for Autonomous Driving
• Autonomous Driving with Deep Learning: A Survey of State-of-Art Technologies
• f-CNN${}^{\text{x}}$: A Toolflow for Mapping Multi-CNN Applications on FPGAs
• Rethinking Self-driving: Multi-task Knowledge for Better Generalization and Accident Explanation Ability
• Output Reachable Set Estimation and Verification for Multi-Layer Neural Networks
• An embedded system for the automated generation of labeled plant images to enable machine learning applications in agriculture
• Key Ingredients of Self-Driving Cars
• Learning from Maps: Visual Common Sense for Autonomous Driving
• DeepRacer: Educational Autonomous Racing Platform for Experimentation with Sim2Real Reinforcement Learning
• Deep Imitation Learning for Autonomous Driving in Generic Urban Scenarios with Enhanced Safety
• UPDeT: Universal Multi-agent Reinforcement Learning via Policy Decoupling with Transformers
• DeepTerramechanics: Terrain Classification and Slip Estimation for Ground Robots via Deep Learning
• Goal-conditioned Imitation Learning
• Aggressive Deep Driving: Model Predictive Control with a CNN Cost Model
• Imitation Learning via Off-Policy Distribution Matching
• Model-free Deep Reinforcement Learning for Urban Autonomous Driving
• Budgeted Training: Rethinking Deep Neural Network Training Under Resource Constraints
• How hard is it to cross the room? -- Training (Recurrent) Neural Networks to steer a UAV
• Dynamic Conditional Imitation Learning for Autonomous Driving
• Overview of Tools Supporting Planning for Automated Driving
• Glance and Focus: a Dynamic Approach to Reducing Spatial Redundancy in Image Classification
• Opening the black box of neural nets: case studies in stop/top discrimination
• Fortified Networks: Improving the Robustness of Deep Networks by Modeling the Manifold of Hidden Representations
• Paracosm: A Language and Tool for Testing Autonomous Driving Systems
• Dynamic Adversarial Patch for Evading Object Detection Models
• Opportunities and Challenges in Deep Learning Adversarial Robustness: A Survey
• Outside the Box: Abstraction-Based Monitoring of Neural Networks
• Towards Safe Machine Learning for CPS: Infer Uncertainty from Training Data
• Learning to Solve Network Flow Problems via Neural Decoding
• Learning and Communications Co-Design for Remote Inference Systems: Feature Length Selection and Transmission Scheduling
• AI Research Considerations for Human Existential Safety (ARCHES)
• Interpolated Adversarial Training: Achieving Robust Neural Networks without Sacrificing Too Much Accuracy
• Autonomous Driving in Reality with Reinforcement Learning and Image Translation
• Learning to Reach Goals via Iterated Supervised Learning
• Human-Centered Autonomous Vehicle Systems: Principles of Effective Shared Autonomy
• Organization of machine learning based product development as per ISO 26262 and ISO/PAS 21448
• Rogue Signs: Deceiving Traffic Sign Recognition with Malicious Ads and Logos
• Operational Calibration: Debugging Confidence Errors for DNNs in the Field
• Computing Systems for Autonomous Driving: State-of-the-Art and Challenges
• Interpretable End-to-end Urban Autonomous Driving with Latent Deep Reinforcement Learning
• TrISec: Training Data-Unaware Imperceptible Security Attacks on Deep Neural Networks
• TrojDRL: Trojan Attacks on Deep Reinforcement Learning Agents
• In Defense of Classical Image Processing: Fast Depth Completion on the CPU
• Solving Optimization Problems through Fully Convolutional Networks: an Application to the Travelling Salesman Problem
• Recall Traces: Backtracking Models for Efficient Reinforcement Learning
• Combining Neural Networks and Tree Search for Task and Motion Planning in Challenging Environments
• Software Engineering Practice in the Development of Deep Learning Applications
• Learning Self-Imitating Diverse Policies
• Disentangling Label Distribution for Long-tailed Visual Recognition
• Universal adversarial examples in speech command classification
• Advances in centerline estimation for autonomous lateral control
• Inverse reinforcement learning for video games
• Fast Neural Network Verification via Shadow Prices
• Towards Compact and Robust Deep Neural Networks
• FRENETIX: A High-Performance and Modular Motion Planning Framework for Autonomous Driving
• Visual Imitation Made Easy
• Once-for-All Adversarial Training: In-Situ Tradeoff between Robustness and Accuracy for Free
• A Tour of Reinforcement Learning: The View from Continuous Control
• Deep Reinforcement Learning for Human-Like Driving Policies in Collision Avoidance Tasks of Self-Driving Cars
• Adaptive MPC for Autonomous Lane Keeping
• Rules of the Road: Predicting Driving Behavior with a Convolutional Model of Semantic Interactions
• Self-driving scale car trained by Deep reinforcement learning
• Operation is the hardest teacher: estimating DNN accuracy looking for mispredictions
• Boundary State Generation for Testing and Improvement of Autonomous Driving Systems
• Towards Imperceptible and Robust Adversarial Example Attacks against Neural Networks
• A Behavioral Approach to Visual Navigation with Graph Localization Networks
• Value Iteration Networks on Multiple Levels of Abstraction
• There is Limited Correlation between Coverage and Robustness for Deep Neural Networks
• Adversarial Neural Pruning with Latent Vulnerability Suppression
• Learning Accurate, Comfortable and Human-like Driving
• Road Scene Graph: A Semantic Graph-Based Scene Representation Dataset for Intelligent Vehicles
• Feature Distillation: DNN-Oriented JPEG Compression Against Adversarial Examples
• Agent-Based Simulations of Blockchain protocols illustrated via Kadena's Chainweb
• Marvel: A Data-centric Compiler for DNN Operators on Spatial Accelerators
• Automated Lane Change Strategy using Proximal Policy Optimization-based Deep Reinforcement Learning
• Imitation Learning for High Precision Peg-in-Hole Tasks
• Evolutionary Synthesis of Deep Neural Networks via Synaptic Cluster-driven Genetic Encoding
• A Survey on Autonomous Vehicle Control in the Era of Mixed-Autonomy: From Physics-Based to AI-Guided Driving Policy Learning
• Neural Ensemble Search for Uncertainty Estimation and Dataset Shift
• Robotic Table Tennis: A Case Study into a High Speed Learning System
• Defending Against Adversarial Examples with K-Nearest Neighbor
• Feedback in Imitation Learning: The Three Regimes of Covariate Shift
• Improving Interpretability of Deep Neural Networks with Semantic Information
• Attended Temperature Scaling: A Practical Approach for Calibrating Deep Neural Networks
• Feature Re-Learning with Data Augmentation for Video Relevance Prediction
• Rethinking Semantic Segmentation Evaluation for Explainability and Model Selection
• Learning to Generalize Across Long-Horizon Tasks from Human Demonstrations
• Toward Simulating Environments in Reinforcement Learning Based Recommendations
• Single Shot MC Dropout Approximation
• Vision-Language Navigation with Self-Supervised Auxiliary Reasoning Tasks
• Simple Physical Adversarial Examples against End-to-End Autonomous Driving Models
• CopyCAT: Taking Control of Neural Policies with Constant Attacks
• Deep Model Predictive Control with Stability Guarantees
• PLOP: Probabilistic poLynomial Objects trajectory Planning for autonomous driving
• Intermediate Deep Feature Compression: the Next Battlefield of Intelligent Sensing
• On Evaluation of Adversarial Perturbations for Sequence-to-Sequence Models
• Coverage Testing of Deep Learning Models using Dataset Characterization
• Predicting Model Failure using Saliency Maps in Autonomous Driving Systems
• Quantum Gradient Algorithm for General Polynomials
• Spatial Uncertainty Sampling for End-to-End Control
• Confidence Scoring Using Whitebox Meta-models with Linear Classifier Probes
• Is it Safe to Drive? An Overview of Factors, Challenges, and Datasets for Driveability Assessment in Autonomous Driving
• An Auto-tuning Framework for Autonomous Vehicles
• Moral Dilemmas for Moral Machines
• Road Context-aware Intrusion Detection System for Autonomous Cars
• DeepObfuscation: Securing the Structure of Convolutional Neural Networks via Knowledge Distillation
• A Hierarchical Architecture for Sequential Decision-Making in Autonomous Driving using Deep Reinforcement Learning
• Learning Interaction-aware Guidance Policies for Motion Planning in Dense Traffic Scenarios
• Cycle-of-Learning for Autonomous Systems from Human Interaction
• Transferring Autonomous Driving Knowledge on Simulated and Real Intersections
• Verifying Aircraft Collision Avoidance Neural Networks Through Linear Approximations of Safe Regions
• Textual Explanations for Self-Driving Vehicles
• SAM: Squeeze-and-Mimic Networks for Conditional Visual Driving Policy Learning
• Learning to Steer by Mimicking Features from Heterogeneous Auxiliary Networks
• Accelerating High-Strain Continuum-Scale Brittle Fracture Simulations with Machine Learning
• A Little Fog for a Large Turn
• Towards Quantification of Assurance for Learning-enabled Components
• An Analysis of Adversarial Attacks and Defenses on Autonomous Driving Models
• Explaining Explanations to Society
• Human-Inspired Multi-Agent Navigation using Knowledge Distillation
• Assisting Blind People Using Object Detection with Vocal Feedback
• CARLA Real Traffic Scenarios -- novel training ground and benchmark for autonomous driving
• Project AutoVision: Localization and 3D Scene Perception for an Autonomous Vehicle with a Multi-Camera System
• Gradient-free Policy Architecture Search and Adaptation
• The NVIDIA PilotNet Experiments
• Handover Control in Wireless Systems via Asynchronous Multi-User Deep Reinforcement Learning
• The Importance of Balanced Data Sets: Analyzing a Vehicle Trajectory Prediction Model based on Neural Networks and Distributed Representations
• DSDNet: Deep Structured self-Driving Network
• PIVEN: A Deep Neural Network for Prediction Intervals with Specific Value Prediction
• ObserveNet Control: A Vision-Dynamics Learning Approach to Predictive Control in Autonomous Vehicles
• PatchAttack: A Black-box Texture-based Attack with Reinforcement Learning
• Enhanced Behavioral Cloning Based self-driving Car Using Transfer Learning
• Fixed-Dimensional and Permutation Invariant State Representation of Autonomous Driving
• Intervention Aided Reinforcement Learning for Safe and Practical Policy Optimization in Navigation
• Lane Change Decision-making through Deep Reinforcement Learning with Rule-based Constraints
• Explaining the Unexplained: A CLass-Enhanced Attentive Response (CLEAR) Approach to Understanding Deep Neural Networks
• Unified Classification and Rejection: A One-versus-All Framework
• Brain Inspired Cognitive Model with Attention for Self-Driving Cars
• DeepGini: Prioritizing Massive Tests to Enhance the Robustness of Deep Neural Networks
• Automated vehicle's behavior decision making using deep reinforcement learning and high-fidelity simulation environment
• TOP: Backdoor Detection in Neural Networks via Transferability of Perturbation
• Counterexample-Guided Learning of Monotonic Neural Networks
• Analyzing Knowledge Transfer in Deep Q-Networks for Autonomously Handling Multiple Intersections
• Hardware-Efficient Photonic Tensor Core: Accelerating Deep Neural Networks with Structured Compression
• Safe-visor Architecture for Sandboxing (AI-based) Unverified Controllers in Stochastic Cyber-Physical Systems
• A New Approach for Resource Scheduling with Deep Reinforcement Learning
• Imitation Learning for End to End Vehicle Longitudinal Control with Forward Camera
• Microscopic Traffic Simulation by Cooperative Multi-agent Deep Reinforcement Learning
• Deep Evidential Regression
• Towards End-to-End Deep Learning for Autonomous Racing: On Data Collection and a Unified Architecture for Steering and Throttle Prediction
• Case Study: Verifying the Safety of an Autonomous Racing Car with a Neural Network Controller
• C^3Net: End-to-End deep learning for efficient real-time visual active camera control
• Who Make Drivers Stop? Towards Driver-centric Risk Assessment: Risk Object Identification via Causal Inference
• Pontryagin Differentiable Programming: An End-to-End Learning and Control Framework
• OctoPath: An OcTree Based Self-Supervised Learning Approach to Local Trajectory Planning for Mobile Robots
• Conditional Driving from Natural Language Instructions
• Learning Awareness Models
• Incorporating Unlabeled Data into Distributionally Robust Learning
• See, Attend and Brake: An Attention-based Saliency Map Prediction Model for End-to-End Driving
• Two-stream convolutional networks for end-to-end learning of self-driving cars
• Compositional ADAM: An Adaptive Compositional Solver
• Autonomous Six-Degree-of-Freedom Spacecraft Docking Maneuvers via Reinforcement Learning
• Cortex: A Compiler for Recursive Deep Learning Models
• A Survey of Behavior Learning Applications in Robotics -- State of the Art and Perspectives
• Responsible AI: Gender bias assessment in emotion recognition
• Snooping Attacks on Deep Reinforcement Learning
• Deep Reinforcement Learning Based High-level Driving Behavior Decision-making Model in Heterogeneous Traffic
• Are all outliers alike? On Understanding the Diversity of Outliers for Detecting OODs
• Attacking Vision-based Perception in End-to-End Autonomous Driving Models
• LIDAR-based Driving Path Generation Using Fully Convolutional Neural Networks
• Natural Language Generation with Neural Variational Models
• AdvSim: Generating Safety-Critical Scenarios for Self-Driving Vehicles
• A Review of Visual Odometry Methods and Its Applications for Autonomous Driving
• Design Space of Behaviour Planning for Autonomous Driving
• End-to-end driving simulation via angle branched network
• Find Your Own Way: Weakly-Supervised Segmentation of Path Proposals for Urban Autonomy
• Grounded Relational Inference: Domain Knowledge Driven Explainable Autonomous Driving
• Recurrent Predictive State Policy Networks
• On the Road with 16 Neurons: Mental Imagery with Bio-inspired Deep Neural Networks
• Understanding Multi-Modal Perception Using Behavioral Cloning for Peg-In-a-Hole Insertion Tasks
• Block Switching: A Stochastic Approach for Deep Learning Security
• Conformal Predictive Safety Filter for RL Controllers in Dynamic Environments
• Affordance-based Reinforcement Learning for Urban Driving
• Improving Calibration for Long-Tailed Recognition
• Learning predictive representations in autonomous driving to improve deep reinforcement learning
• Certifying Confidence via Randomized Smoothing
• Should Ensemble Members Be Calibrated?
• Triple Generative Adversarial Networks
• Revisiting One-vs-All Classifiers for Predictive Uncertainty and Out-of-Distribution Detection in Neural Networks
• BayesRace: Learning to race autonomously using prior experience
• Improving the Generalization of End-to-End Driving through Procedural Generation
• Automating Vehicles by Deep Reinforcement Learning using Task Separation with Hill Climbing
• Detecting OODs as datapoints with High Uncertainty
• Efficient Learning of Distributed Linear-Quadratic Controllers
• Learning-based Initialization of Trajectory Optimization for Path-following Problems of Redundant Manipulators
• Jacobian Adversarially Regularized Networks for Robustness
• Targeted Nonlinear Adversarial Perturbations in Images and Videos
• Using Undervolting as an On-Device Defense Against Adversarial Machine Learning Attacks
• Soft Calibration Objectives for Neural Networks
• Wolf in Sheep's Clothing - The Downscaling Attack Against Deep Learning Applications
• TACAN: Transmitter Authentication through Covert Channels in Controller Area Networks
• Guided Policy Search Model-based Reinforcement Learning for Urban Autonomous Driving
• Affordance Learning In Direct Perception for Autonomous Driving
• Explorations and Lessons Learned in Building an Autonomous Formula SAE Car from Simulations
• Multi-Class Gaussian Process Classification Made Conjugate: Efficient Inference via Data Augmentation
• Diversity regularization in deep ensembles
• DeepRacing: Parameterized Trajectories for Autonomous Racing
• Robust Deep Sensing Through Transfer Learning in Cognitive Radio
• Meta learning Framework for Automated Driving
• Comparing Apples and Oranges: Off-Road Pedestrian Detection on the NREC Agricultural Person-Detection Dataset
• Real-time End-to-End Federated Learning: An Automotive Case Study
• Learning hierarchical behavior and motion planning for autonomous driving
• Dangers of Bayesian Model Averaging under Covariate Shift
• Feature Boosting, Suppression, and Diversification for Fine-Grained Visual Classification
• Decision-Making Technology for Autonomous Vehicles Learning-Based Methods, Applications and Future Outlook
• Latent Space Reinforcement Learning for Steering Angle Prediction
• Sound and Complete Neural Network Repair with Minimality and Locality Guarantees
• Rethinking Recurrent Neural Networks and Other Improvements for Image Classification
• Characterising Across-Stack Optimisations for Deep Convolutional Neural Networks
• Exploiting All Samples in Low-Resource Sentence Classification: Early Stopping and Initialization Parameters
• MeLIME: Meaningful Local Explanation for Machine Learning Models
• Exploring the Role of Machine Learning in Scientific Workflows: Opportunities and Challenges
• ClickBAIT: Click-based Accelerated Incremental Training of Convolutional Neural Networks
• Ara: A 1 GHz+ Scalable and Energy-Efficient RISC-V Vector Processor with Multi-Precision Floating Point Support in 22 nm FD-SOI
• Shift-BNN: Highly-Efficient Probabilistic Bayesian Neural Network Training via Memory-Friendly Pattern Retrieving
• Emergent Road Rules In Multi-Agent Driving Environments
• FaceHack: Triggering backdoored facial recognition systems using facial characteristics
• An Intelligent Safety System for Human-Centered Semi-Autonomous Vehicles
• Fighting Failures with FIRE: Failure Identification to Reduce Expert Burden in Intervention-Based Learning
• Transformers for One-Shot Visual Imitation
• Exploring Uncertainty in Deep Learning for Construction of Prediction Intervals
• Mitigating backdoor attacks in LSTM-based Text Classification Systems by Backdoor Keyword Identification
• Policy Learning Using Weak Supervision
• Detection Defense Against Adversarial Attacks with Saliency Map
• Vision-Based Autonomous Car Racing Using Deep Imitative Reinforcement Learning
• A Survey of Deep Reinforcement Learning Algorithms for Motion Planning and Control of Autonomous Vehicles
• A Tale of Two Approximations: Tightening Over-Approximation for DNN Robustness Verification via Under-Approximation
• DeepRT: A Soft Real Time Scheduler for Computer Vision Applications on the Edge
• Learning a CNN-based End-to-End Controller for a Formula SAE Racecar
• Reinforcement Learning Based Safe Decision Making for Highway Autonomous Driving
• Shaping Rewards for Reinforcement Learning with Imperfect Demonstrations using Generative Models
• Closing the gap towards end-to-end autonomous vehicle system
• Explainable Object-induced Action Decision for Autonomous Vehicles
• Deep Object-Centric Policies for Autonomous Driving
• HoneyModels: Machine Learning Honeypots
• Hierarchical Policy for Non-prehensile Multi-object Rearrangement with Deep Reinforcement Learning and Monte Carlo Tree Search
• DREAM Architecture: a Developmental Approach to Open-Ended Learning in Robotics
• Maximum Resilience of Artificial Neural Networks
• Certified Control: An Architecture for Verifiable Safety of Autonomous Vehicles
• Reach-SDP: Reachability Analysis of Closed-Loop Systems with Neural Network Controllers via Semidefinite Programming
• Adversarial Imitation Learning from Incomplete Demonstrations
• Evolutionary Multi-objective Optimisation in Neurotrajectory Prediction
• Model-based Behavioral Cloning with Future Image Similarity Learning
• Agile Autonomous Driving using End-to-End Deep Imitation Learning
• Deep Learning in the Automotive Industry: Recent Advances and Application Examples
• Periphery-Fovea Multi-Resolution Driving Model guided by Human Attention
• HG-Caffe: Mobile and Embedded Neural Network GPU (OpenCL) Inference Engine with FP16 Supporting
• Imitating Driver Behavior with Generative Adversarial Networks
• Cooperative Highway Work Zone Merge Control based on Reinforcement Learning in A Connected and Automated Environment
• Deep Learning and Control Algorithms of Direct Perception for Autonomous Driving
• Scene-Graph Augmented Data-Driven Risk Assessment of Autonomous Vehicle Decisions
• Flexible Deep Neural Network Processing
• Reinforcement Learning based Control of Imitative Policies for Near-Accident Driving
• Improving Robustness of Learning-based Autonomous Steering Using Adversarial Images
• Environment Predictive Coding for Embodied Agents
• A Survey on Understanding, Visualizations, and Explanation of Deep Neural Networks
• Towards Calibrated Model for Long-Tailed Visual Recognition from Prior Perspective
• AI based Algorithms of Path Planning, Navigation and Control for Mobile Ground Robots and UAVs
• Modular Vehicle Control for Transferring Semantic Information Between Weather Conditions Using GANs
• Continuous Risk Measures for Driving Support
• Verification of Binarized Neural Networks via Inter-Neuron Factoring
• Multi Agent Navigation in Unconstrained Environments using a Centralized Attention based Graphical Neural Network Controller
• OVERT: An Algorithm for Safety Verification of Neural Network Control Policies for Nonlinear Systems
• PAC Prediction Sets Under Covariate Shift
• Seeing All the Angles: Learning Multiview Manipulation Policies for Contact-Rich Tasks from Demonstrations
• Generalizing Across Multi-Objective Reward Functions in Deep Reinforcement Learning
• SOON: Scenario Oriented Object Navigation with Graph-based Exploration
• Action-Based Representation Learning for Autonomous Driving
• Adversarial Learning with Margin-based Triplet Embedding Regularization
• Automated Lane Change Decision Making using Deep Reinforcement Learning in Dynamic and Uncertain Highway Environment
• Reachable Set Estimation for Neural Network Control Systems: A Simulation-Guided Approach
• Playing Minecraft with Behavioural Cloning
• Rethinking Performance Estimation in Neural Architecture Search
• Vehicle trajectory prediction in top-view image sequences based on deep learning method
• Importance Resampling for Off-policy Prediction
• Graph Neural Lasso for Dynamic Network Regression
• Federated Imitation Learning: A Privacy Considered Imitation Learning Framework for Cloud Robotic Systems with Heterogeneous Sensor Data
• Human-in-the-Loop Methods for Data-Driven and Reinforcement Learning Systems
• LOCAL: Low-Complex Mapping Algorithm for Spatial DNN Accelerators
• AutoSoC: Automating Algorithm-SOC Co-design for Aerial Robots
• On Machine Learning and Structure for Mobile Robots
• Behavioral Research and Practical Models of Drivers' Attention
• MultiNet: Multi-Modal Multi-Task Learning for Autonomous Driving
• Improving Robustness using Generated Data
• SAFARI: Safe and Active Robot Imitation Learning with Imagination
• Formal Verification of Robustness and Resilience of Learning-Enabled State Estimation Systems
• End-to-end Interpretable Neural Motion Planner
• Learning by Cheating
• TrafficSim: Learning to Simulate Realistic Multi-Agent Behaviors
• Neuro-Symbolic Execution: The Feasibility of an Inductive Approach to Symbolic Execution
• Driving Experience Transfer Method for End-to-End Control of Self-Driving Cars
• Preventing Imitation Learning with Adversarial Policy Ensembles
• Melodic Phrase Segmentation By Deep Neural Networks
• Pruning and Slicing Neural Networks using Formal Verification
• mFI-PSO: A Flexible and Effective Method in Adversarial Image Generation for Deep Neural Networks
• A Systematic Comparison of Deep Learning Architectures in an Autonomous Vehicle
• A Review of Formal Methods applied to Machine Learning
• Multi-task Learning with Attention for End-to-end Autonomous Driving
• Feedback from Pixels: Output Regulation via Learning-Based Scene View Synthesis
• ASP:A Fast Adversarial Attack Example Generation Framework based on Adversarial Saliency Prediction
• Multi-modal Experts Network for Autonomous Driving
• Differentially Private Adversarial Robustness Through Randomized Perturbations
• Recurrent Neural Network-based Model for Accelerated Trajectory Analysis in AIMD Simulations
• Hierarchical Multi-task Deep Neural Network Architecture for End-to-End Driving
• Avoidance of Manual Labeling in Robotic Autonomous Navigation Through Multi-Sensory Semi-Supervised Learning
• On Value Discrepancy of Imitation Learning
• Teaching UAVs to Race: End-to-End Regression of Agile Controls in Simulation
• Continual Learning: Tackling Catastrophic Forgetting in Deep Neural Networks with Replay Processes
• Feature Analysis and Selection for Training an End-to-End Autonomous Vehicle Controller Using the Deep Learning Approach
• DeepFault: Fault Localization for Deep Neural Networks
• A Methodological Review of Visual Road Recognition Procedures for Autonomous Driving Applications
• Unsupervised Temperature Scaling: An Unsupervised Post-Processing Calibration Method of Deep Networks
• A Polynomial-Based Approach for Architectural Design and Learning with Deep Neural Networks
• Debona: Decoupled Boundary Network Analysis for Tighter Bounds and Faster Adversarial Robustness Proofs
• Accelerated Convolutions for Efficient Multi-Scale Time to Contact Computation in Julia
• Vision-based Navigation of Autonomous Vehicle in Roadway Environments with Unexpected Hazards
• Vision-Dialog Navigation by Exploring Cross-modal Memory
• AutoPilot: Automating SoC Design Space Exploration for SWaP Constrained Autonomous UAVs
• DRIBO: Robust Deep Reinforcement Learning via Multi-View Information Bottleneck
• A machine learning environment for evaluating autonomous driving software
• Defending Against Image Corruptions Through Adversarial Augmentations
• Policy Smoothing for Provably Robust Reinforcement Learning
• PRED18: Dataset and Further Experiments with DAVIS Event Camera in Predator-Prey Robot Chasing
• Probably Approximately Correct Vision-Based Planning using Motion Primitives
• Roofline Model for UAVs:A Bottleneck Analysis Tool for Designing Compute Systems for Autonomous Drones
• Exploring Covariate and Concept Shift for Detection and Calibration of Out-of-Distribution Data
• A Mapping of Assurance Techniques for Learning Enabled Autonomous Systems to the Systems Engineering Lifecycle
• Defining the problem of Observation Learning
• CIRL: Controllable Imitative Reinforcement Learning for Vision-based Self-driving
• A universal duplication-free quantum neural network
• Adaptive Feature Alignment for Adversarial Training
• Probabilistic Verification of Neural Networks Against Group Fairness
• Neural Network Branch-and-Bound for Neural Network Verification
• WAD: A Deep Reinforcement Learning Agent for Urban Autonomous Driving
• Ignition: An End-to-End Supervised Model for Training Simulated Self-Driving Vehicles
• ML + FV = $♡$? A Survey on the Application of Machine Learning to Formal Verification
• VisualBackProp for learning using privileged information with CNNs
• Simulation Environment for Safety Assessment of CEAV Deployment in Linden
• Deep Latent Competition: Learning to Race Using Visual Control Policies in Latent Space
• Investigating Behavior Cloning from Few Demonstrations for Autonomous Driving based on Bird's-Eye View in Simulated Cities
• Learning Task-specific Representation for Novel Words in Sequence Labeling
• Deep Learning Based Motion Planning For Autonomous Vehicle Using Spatiotemporal LSTM Network
• Learning On-Road Visual Control for Self-Driving Vehicles with Auxiliary Tasks
• End-to-End Vision-Based Adaptive Cruise Control (ACC) Using Deep Reinforcement Learning
• Automated Analysis of Femoral Artery Calcification Using Machine Learning Techniques
• An Adaptive View of Adversarial Robustness from Test-time Smoothing Defense
• DeepSmartFuzzer: Reward Guided Test Generation For Deep Learning
• Winning the ICCV 2019 Learning to Drive Challenge
• Evolutionary Selective Imitation: Interpretable Agents by Imitation Learning Without a Demonstrator
• Defining Digital Quadruplets in the Cyber-Physical-Social Space for Parallel Driving
• Live Trojan Attacks on Deep Neural Networks
• On the Potential of Smarter Multi-layer Maps
• DDD20 End-to-End Event Camera Driving Dataset: Fusing Frames and Events with Deep Learning for Improved Steering Prediction
• Spirit Distillation: Precise Real-time Semantic Segmentation of Road Scenes with Insufficient Data
• Adversarial Imitation Learning with Trajectorial Augmentation and Correction
• RISC-NN: Use RISC, NOT CISC as Neural Network Hardware Infrastructure
• Advances in Inference and Representation for Simultaneous Localization and Mapping
• Decision-Making under On-Ramp merge Scenarios by Distributional Soft Actor-Critic Algorithm
• Approaching Neural Network Uncertainty Realism
• A Novel Regression Loss for Non-Parametric Uncertainty Optimization
• ROBY: Evaluating the Robustness of a Deep Model by its Decision Boundaries
• Cross-domain Imitation from Observations
• Enabling the Evaluation of Driver Physiology Via Vehicle Dynamics
• Decentralized and Asymmetric Multi-Agent Learning in Construction Sites
• Programmatic Imitation Learning from Unlabeled and Noisy Demonstrations
• A Survey of FPGA-Based Robotic Computing
• SOCRATES: Towards a Unified Platform for Neural Network Analysis
• Interpretable Self-Attention Temporal Reasoning for Driving Behavior Understanding
• Towards Interpreting Recurrent Neural Networks through Probabilistic Abstraction
• AIBA: An AI Model for Behavior Arbitration in Autonomous Driving
• Zero-shot Deep Reinforcement Learning Driving Policy Transfer for Autonomous Vehicles based on Robust Control
• Developing a Bubble Chamber Particle Discriminator Using Semi-Supervised Learning
• nn-dependability-kit: Engineering Neural Networks for Safety-Critical Autonomous Driving Systems
• 3D Traffic Simulation for Autonomous Vehicles in Unity and Python
• Dynamic Risk Assessment for Vehicles of Higher Automation Levels by Deep Learning
• Bayesian Prediction of Future Street Scenes through Importance Sampling based Optimization
• Quantitative Projection Coverage for Testing ML-enabled Autonomous Systems
• Hybrid Piezoelectric-Magnetic Neurons: A Proposal for Energy-Efficient Machine Learning
• Data Analytics Service Composition and Deployment on Edge Devices
• Efficient and Robust LiDAR-Based End-to-End Navigation
• Practical Convex Formulation of Robust One-hidden-layer Neural Network Training
• A First Look at Mobile Intelligence: Architecture, Experimentation and Challenges
• What data do we need for training an AV motion planner?
• A perception centred self-driving system without HD Maps
• Learning Negotiating Behavior Between Cars in Intersections using Deep Q-Learning
• Clipping free attacks against artificial neural networks
• Brain-Inspired Deep Imitation Learning for Autonomous Driving Systems
• Meta-Adversarial Inverse Reinforcement Learning for Decision-making Tasks
• Learning Resilient Behaviors for Navigation Under Uncertainty
• Iterative Imitation Policy Improvement for Interactive Autonomous Driving
• Learning Neural Models for Natural Language Processing in the Face of Distributional Shift
• Solving an elastic inverse problem using Convolutional Neural Networks
• Attentional Bottleneck: Towards an Interpretable Deep Driving Network
• Reachable Set Estimation and Verification for Neural Network Models of Nonlinear Dynamic Systems
• Learning Calibratable Policies using Programmatic Style-Consistency
• TASO: Time and Space Optimization for Memory-Constrained DNN Inference
• ADAPS: Autonomous Driving Via Principled Simulations
• Better AI through Logical Scaffolding
• Reparameterized Variational Divergence Minimization for Stable Imitation
• Affordable Modular Autonomous Vehicle Development Platform
• Learning Interpretable End-to-End Vision-Based Motion Planning for Autonomous Driving with Optical Flow Distillation
• Generating Adversarial Inputs Using A Black-box Differential Technique
• Safe learning-based optimal motion planning for automated driving
• Adma: A Flexible Loss Function for Neural Networks
• Dynamic Defense Approach for Adversarial Robustness in Deep Neural Networks via Stochastic Ensemble Smoothed Model
• Black-box Adversarial Sample Generation Based on Differential Evolution
• Driving among Flatmobiles: Bird-Eye-View occupancy grids from a monocular camera for holistic trajectory planning
• Deep Reinforcement Learning with Mixed Convolutional Network
• Unifying Variational Inference and PAC-Bayes for Supervised Learning that Scales
• Long-Term On-Board Prediction of People in Traffic Scenes under Uncertainty
• Sensor Fusion: Gated Recurrent Fusion to Learn Driving Behavior from Temporal Multimodal Data
• Requirements for Developing Robust Neural Networks
• Fighting Copycat Agents in Behavioral Cloning from Observation Histories
• Multi-Vehicle Interaction Scenarios Generation with Interpretable Traffic Primitives and Gaussian Process Regression
• Using Segmentation Masks in the ICCV 2019 Learning to Drive Challenge
• Long-term Prediction of Vehicle Behavior using Short-term Uncertainty-aware Trajectories and High-definition Maps
• Fast-UAP: An Algorithm for Speeding up Universal Adversarial Perturbation Generation with Orientation of Perturbation Vectors
• VUNet: Dynamic Scene View Synthesis for Traversability Estimation using an RGB Camera
• AReN: Assured ReLU NN Architecture for Model Predictive Control of LTI Systems
• Grounding Human-to-Vehicle Advice for Self-driving Vehicles
• Accurate Trajectory Prediction for Autonomous Vehicles
• N 2 C : Neural Network Controller Design Using Behavioral Cloning
• Tolerance-Guided Policy Learning for Adaptable and Transferrable Delicate Industrial Insertion
• DeepGoal: Learning to Drive with driving intention from Human Control Demonstration
• Simplifying Neural Networks using Formal Verification
• Feudal Steering: Hierarchical Learning for Steering Angle Prediction
• Aggressive Perception-Aware Navigation using Deep Optical Flow Dynamics and PixelMPC
• Detecting Adversarial Samples Using Density Ratio Estimates
• Learning to drive via Apprenticeship Learning and Deep Reinforcement Learning
• Efficient Latent Representations using Multiple Tasks for Autonomous Driving
• Parallelized Interactive Machine Learning on Autonomous Vehicles
• Deformation Robust Roto-Scale-Translation Equivariant CNNs
• Behavior Planning at Urban Intersections through Hierarchical Reinforcement Learning
• Episodic Self-Imitation Learning with Hindsight
• Vehicle Trajectory Prediction by Transfer Learning of Semi-Supervised Models
• Investigation of Factorized Optical Flows as Mid-Level Representations
• Deep Structured Reactive Planning
• A Safe Hierarchical Planning Framework for Complex Driving Scenarios based on Reinforcement Learning
• Perceptual Attention-based Predictive Control
• RTGPU: Real-Time GPU Scheduling of Hard Deadline Parallel Tasks with Fine-Grain Utilization
• Human Correspondence Consensus for 3D Object Semantic Understanding
• An Online Evolving Framework for Modeling the Safe Autonomous Vehicle Control System via Online Recognition of Latent Risks
• Predicting Real-time Scientific Experiments Using Transformer models and Reinforcement Learning
• Instance-Aware Predictive Navigation in Multi-Agent Environments
• Optimizing the Whole-life Cost in End-to-end CNN Acceleration
• An Empirical Investigation of Randomized Defenses against Adversarial Attacks
• PODNet: A Neural Network for Discovery of Plannable Options
• Deep Minimax Probability Machine
• A Novel Unsupervised Post-Processing Calibration Method for DNNS with Robustness to Domain Shift
• Covert Channel-Based Transmitter Authentication in Controller Area Networks
• Relational Mimic for Visual Adversarial Imitation Learning
• ReNeg and Backseat Driver: Learning from Demonstration with Continuous Human Feedback
• Autonomous Cars: Vision based Steering Wheel Angle Estimation
• Learning to Predict Ego-Vehicle Poses for Sampling-Based Nonholonomic Motion Planning
• Can Intelligent Hyperparameter Selection Improve Resistance to Adversarial Examples?
• RT-Gang: Real-Time Gang Scheduling Framework for Safety-Critical Systems
• Learning to Navigate Unseen Environments: Back Translation with Environmental Dropout
• Single-step Options for Adversary Driving
• Integration of Imitation Learning using GAIL and Reinforcement Learning using Task-achievement Rewards via Probabilistic Graphical Model
• Utilizing Eye Gaze to Enhance the Generalization of Imitation Networks to Unseen Environments
• Context Aware Road-user Importance Estimation (iCARE)
• Asynchronous "Events" are Better For Motion Estimation
• Safer End-to-End Autonomous Driving via Conditional Imitation Learning and Command Augmentation
• A hybrid controller for safe and efficient collision avoidance control
• Convolutional nets for reconstructing neural circuits from brain images acquired by serial section electron microscopy
• Time-Dependent Hybrid-State A* and Optimal Control for Autonomous Vehicles in Arbitrary and Dynamic Environment
• Meta Adaptation using Importance Weighted Demonstrations
• Adversarial Learning-Based On-Line Anomaly Monitoring for Assured Autonomy
• Uncertainty-Aware Data Aggregation for Deep Imitation Learning
• A Modeled Approach for Online Adversarial Test of Operational Vehicle Safety (extended version)
• Imitation Learning of Hierarchical Driving Model: from Continuous Intention to Continuous Trajectory
• Large scale visual place recognition with sub-linear storage growth
• On the Guaranteed Almost Equivalence between Imitation Learning from Observation and Demonstration
• A Simple Framework to Quantify Different Types of Uncertainty in Deep Neural Networks for Image Classification
• Variational Laplace for Bayesian neural networks
• Novelty Detection via Network Saliency in Visual-based Deep Learning
• Normalization effects on shallow neural networks and related asymptotic expansions
• Of Moments and Matching: A Game-Theoretic Framework for Closing the Imitation Gap
• On Offline Evaluation of Vision-based Driving Models
• Multivariate Deep Evidential Regression
• Knowledge Integrated Classifier Design Based on Utility Optimization
• Car Monitoring System in Apartment Garages by Small Autonomous Car using Deep Learning
• End-to-End Race Driving with Deep Reinforcement Learning
• One Network to Solve Them All: A Sequential Multi-Task Joint Learning Network Framework for MR Imaging Pipeline
• DeepGuard: A Framework for Safeguarding Autonomous Driving Systems from Inconsistent Behavior
• Yaw-Guided Imitation Learning for Autonomous Driving in Urban Environments
• Shared Multi-Task Imitation Learning for Indoor Self-Navigation
• On Scalable Testing of Samplers
• DriveNetBench: An Affordable and Configurable Single-Camera Benchmarking System for Autonomous Driving Networks
• Evolutionary Hyperparameter Optimization to Find Lightweight CNN Models for Autonomous Steering
• Why Calibration Error is Wrong Given Model Uncertainty: Using Posterior Predictive Checks with Deep Learning
• Sidekick Policy Learning for Active Visual Exploration
• Weakly Supervised Learning Creates a Fusion of Modeling Cultures
• Vision-Guided Forecasting -- Visual Context for Multi-Horizon Time Series Forecasting
• Assured Neural Network Architectures for Control and Identification of Nonlinear Systems
• Multi-lane Cruising Using Hierarchical Planning and Reinforcement Learning
• Natural Attribute-based Shift Detection
• Examining the causal structures of deep neural networks using information theory
• All-Weather Object Recognition Using Radar and Infrared Sensing
• NEARL: Non-Explicit Action Reinforcement Learning for Robotic Control
• Sequential Dynamic Decision Making with Deep Neural Nets on a Test-Time Budget
• Modeling Preemptive Behaviors for Uncommon Hazardous Situations From Demonstrations
• The Aleatoric Uncertainty Estimation Using a Separate Formulation with Virtual Residuals
• Defense-guided Transferable Adversarial Attacks
• Neural Rate Control for Video Encoding using Imitation Learning
• First-Person Perceptual Guidance Behavior Decomposition using Active Constraint Classification
• SdcNet: A Computation-Efficient CNN for Object Recognition
• Incremental Verification of Fixed-Point Implementations of Neural Networks
• Analyzing Representations inside Convolutional Neural Networks
• Stochastic Action Prediction for Imitation Learning
• Multi-Instance Aware Localization for End-to-End Imitation Learning
• User-centric Composable Services: A New Generation of Personal Data Analytics
• Context-Aware Image Denoising with Auto-Threshold Canny Edge Detection to Suppress Adversarial Perturbation
• Refinement Type Directed Search for Meta-Interpretive-Learning of Higher-Order Logic Programs
• Self-Checking Deep Neural Networks in Deployment
• Per-Pixel Feedback for improving Semantic Segmentation
• Incorporating Orientations into End-to-end Driving Model for Steering Control
• Learning robust driving policies without online exploration
• Convex Online Video Frame Subset Selection using Multiple Criteria for Data Efficient Autonomous Driving
• No Need for Interactions: Robust Model-Based Imitation Learning using Neural ODE
• Deep Learning-Based Autonomous Driving Systems: A Survey of Attacks and Defenses
• Simpler Certified Radius Maximization by Propagating Covariances
• RoadText-1K: Text Detection & Recognition Dataset for Driving Videos
• Guarded Deep Learning using Scenario-Based Modeling
• An online evolving framework for advancing reinforcement-learning based automated vehicle control
• Imitation Learning Approach for AI Driving Olympics Trained on Real-world and Simulation Data Simultaneously
• DinerDash Gym: A Benchmark for Policy Learning in High-Dimensional Action Space
• Enhanced Transfer Learning for Autonomous Driving with Systematic Accident Simulation
• Learning Navigation by Visual Localization and Trajectory Prediction
• Learning a Safety Verifiable Adaptive Cruise Controller from Human Driving Data