138 citations · 402 across the 159 of their papers we have counts for
33 papers · 1 filter
Probe-Space Preconditioning for Fast and Stable Zero-Order Training
Francois Chaubard, Mykel J. Kochenderfer, Chris Ré
Backpropagation (BP) dominates deep learning but imposes a massive memory tax. For example, training OPT-30B with Adam requires 600GB of GPU memory (assuming batch size 8…
Scaling Zero-Order Pretraining through Model Sharding
Francois Chaubard, Mykel J. Kochenderfer, Chris Ré
Zero-order optimization (ZO) trains without backpropagation, making it relevant to forward-only hardware and non-differentiable loss, but its gradient variance grows with perturbed…
RENEW: Towards Learning World Models and Repairing Model Exploitation from Preferences
Logan Mondal Bhamidipaty, Mykel Kochenderfer, Subramanian Ramamoorthy
World models are widely used in offline reinforcement learning (RL) to improve sample efficiency and generate experience beyond a fixed dataset. However, they are vulnerable to mod…
Simplifying Flow Matching Transformations with Low-Rank Mixture Models
Liam A. Kruse, Houjun Liu, Alexandros E. Tzikas +2
Normalizing flows are powerful generative models that learn an invertible mapping between complex data distributions and simple latent distributions, typically a standard normal de…
Welfare, Improvability, and Variance: A Principal-Agent Approach to Optimal Benchmark Item Aggregation
Andreas Haupt, Justin Hartenstein, Anka Reuel +2
AI benchmarks have well-documented limitations, with prior work examining contamination, saturation, and construct underspecification. Aggregation has received far less attention:…
Mechanistic Interpretability for Learning Assurance of a Vision-Based Landing System
Romeo Valentin, Olivia Beyer Bruvik, Marc R. Schlichting +1
EASA's learning-assurance guidance requires data-driven aviation systems to build and monitor their own situation representation, yet for neural networks the technical means to pro…
Zono-Conformal Prediction: Zonotope-Based Uncertainty Quantification for Regression and Classification Tasks
Laura Lützow, Michael Eichelbeck, Mykel J. Kochenderfer +1
Conformal prediction is a popular uncertainty quantification method that augments a base predictor to return sets of predictions with statistically valid coverage guarantees. Howev…
Scaling Recurrent Neural Networks to a Billion Parameters with Zero-Order Optimization
Francois Chaubard, Mykel Kochenderfer
During inference, Recurrent Neural Networks (RNNs) scale constant in both FLOPs and GPU memory with increasing context length, as they compress all prior tokens into a fixed-size m…
DB-KSVD: Scalable Alternating Optimization for Disentangling High-Dimensional Embedding Spaces
Romeo Valentin, Sydney M. Katz, Vincent Vanhoucke +1
Dictionary learning has recently emerged as a promising approach for mechanistic interpretability of large transformer models. Disentangling high-dimensional transformer embeddings…
Physics-informed Gaussian Processes for Safe Envelope Expansion
D. Isaiah Harp, Joshua Ott, Dylan M. Asmar +2
Flight test analysis often requires predefined test points with arbitrarily tight tolerances, leading to extensive and resource-intensive experimental campaigns. To address this ch…
Beyond Gradient Averaging in Parallel Optimization: Improved Robustness through Gradient Agreement Filtering
Francois Chaubard, Duncan Eddy, Mykel J. Kochenderfer
We introduce Gradient Agreement Filtering (GAF) to improve on gradient averaging in distributed deep learning optimization. Traditional distributed data-parallel stochastic gradien…
Graph Q-Learning for Combinatorial Optimization
Victoria M. Dax, Jiachen Li, Kevin Leahy +1
Graph-structured data is ubiquitous throughout natural and social sciences, and Graph Neural Networks (GNNs) have recently been shown to be effective at solving prediction and infe…
A Holistic Assessment of the Reliability of Machine Learning Systems
Anthony Corso, David Karamadian, Romeo Valentin +2
As machine learning (ML) systems increasingly permeate high-stakes settings such as healthcare, transportation, military, and national security, concerns regarding their reliabilit…
A Deep Reinforcement Learning Approach to Rare Event Estimation
Anthony Corso, Kyu-Young Kim, Shubh Gupta +2
An important step in the design of autonomous systems is to evaluate the probability that a failure will occur. In safety-critical domains, the failure probability is extremely sma…
Agent-Time Attention for Sparse Rewards Multi-Agent Reinforcement Learning
Jennifer She, Jayesh K. Gupta, Mykel J. Kochenderfer
Sparse and delayed rewards pose a challenge to single agent reinforcement learning. This challenge is amplified in multi-agent reinforcement learning (MARL) where credit assignment…
Interaction Modeling with Multiplex Attention
Fan-Yun Sun, Isaac Kauvar, Ruohan Zhang +4
Modeling multi-agent systems requires understanding how agents interact. Such systems are often difficult to model because they can involve a variety of types of interactions that…
Disentangling Epistemic and Aleatoric Uncertainty in Reinforcement Learning
Bertrand Charpentier, Ransalu Senanayake, Mykel Kochenderfer +1
Characterizing aleatoric and epistemic uncertainty on the predicted rewards can help in building reliable reinforcement learning (RL) systems. Aleatoric uncertainty results from th…
Meta-SysId: A Meta-Learning Approach for Simultaneous Identification and Prediction
Junyoung Park, Federico Berto, Arec Jamgochian +2
In this paper, we propose Meta-SysId, a meta-learning approach to model sets of systems that have behavior governed by common but unknown laws and that differentiate themselves by…
A Mixed Integer Programming Approach for Verifying Properties of Binarized Neural Networks
Christopher Lazarus, Mykel J. Kochenderfer
Many approaches for verifying input-output properties of neural networks have been proposed recently. However, existing algorithms do not scale well to large networks. Recent work…
Deep Binary Reinforcement Learning for Scalable Verification
Christopher Lazarus, Mykel J. Kochenderfer
The use of neural networks as function approximators has enabled many advances in reinforcement learning (RL). The generalization power of neural networks combined with advances in…
Recursive Reasoning Graph for Multi-Agent Reinforcement Learning
Xiaobai Ma, David Isele, Jayesh K. Gupta +2
Multi-agent reinforcement learning (MARL) provides an efficient way for simultaneously learning policies for multiple agents interacting with each other. However, in scenarios requ…
A Light-Weight Multi-Objective Asynchronous Hyper-Parameter Optimizer
Gabriel Maher, Stephen Boyd, Mykel Kochenderfer +5
We describe a light-weight yet performant system for hyper-parameter optimization that approximately minimizes an overall scalar cost function that is obtained by combining multipl…
Verifying Inverse Model Neural Networks
Chelsea Sidrane, Sydney Katz, Anthony Corso +1
Inverse problems exist in a wide variety of physical domains from aerospace engineering to medical imaging. The goal is to infer the underlying state from a set of observations. Wh…
Conditional Approximate Normalizing Flows for Joint Multi-Step Probabilistic Forecasting with Application to Electricity Demand
Arec Jamgochian, Di Wu, Kunal Menda +2
Some real-world decision-making problems require making probabilistic forecasts over multiple steps at once. However, methods for probabilistic forecasting may fail to capture corr…