198 citations · 362 across the 22 of their papers we have counts for
23 papers
Internalizing Curriculum Judgment for LLM Reinforcement Fine-Tuning
Han Zheng, Yining Ma, Karthick Gunasekaran +4
In LLM Reinforcement Fine-Tuning (RFT), curriculum learning drives both efficiency and performance. Yet, current methods externalize curriculum judgment via handcrafted heuristics…
SpiderGen: Towards Procedure Generation For Carbon Life Cycle Assessments with Generative AI
Anupama Sitaraman, Bharathan Balaji, Yuvraj Agarwal
Investigating the effects of climate change and global warming caused by GHG emissions have been a key concern worldwide. These emissions are largely contributed to by the producti…
RedunCut: Measurement-Driven Sampling and Accuracy Performance Modeling for Low-Cost Live Video Analytics
Gur-Eyal Sela, Kumar Krishna Agrawal, Bharathan Balaji +2
Live video analytics (LVA) runs continuously across massive camera fleets, but inference cost with modern vision models remains high. To address this, dynamic model size selection…
An Expert-grounded benchmark of General Purpose LLMs in LCA
Artur Donaldson, Bharathan Balaji, Cajetan Oriekezie +2
Purpose: Artificial intelligence (AI), and in particular large language models (LLMs), are increasingly being explored as tools to support life cycle assessment (LCA). While demons…
CF-RAG: A Dataset and Method for Carbon Footprint QA Using Retrieval-Augmented Generation
Kaiwen Zhao, Bharathan Balaji, Stephen Lee
Product sustainability reports provide valuable insights into the environmental impacts of a product and are often distributed in PDF format. These reports often include a combinat…
FUELVISION: A Multimodal Data Fusion and Multimodel Ensemble Algorithm for Wildfire Fuels Mapping
Riyaaz Uddien Shaik, Mohamad Alipour, Eric Rowell +3
Accurate assessment of fuel conditions is a prerequisite for fire ignition and behavior prediction, and risk management. The method proposed herein leverages diverse data sources i…
OptiState: State Estimation of Legged Robots using Gated Networks with Transformer-based Vision and Kalman Filtering
Alexander Schperberg, Yusuke Tanaka, Saviz Mowlavi +3
State estimation for legged robots is challenging due to their highly dynamic motion and limitations imposed by sensor accuracy. By integrating Kalman filtering, optimization, and…
Eagle: End-to-end Deep Reinforcement Learning based Autonomous Control of PTZ Cameras
Sandeep Singh Sandha, Bharathan Balaji, Luis Garcia +1
Existing approaches for autonomous control of pan-tilt-zoom (PTZ) cameras use multiple stages where object detection and localization are performed separately from the control of t…
B2RL: An open-source Dataset for Building Batch Reinforcement Learning
Hsin-Yu Liu, Xiaohan Fu, Bharathan Balaji +2
Batch reinforcement learning (BRL) is an emerging research area in the RL community. It learns exclusively from static datasets (i.e. replay buffers) without interaction with the e…
Context-Aware Streaming Perception in Dynamic Environments
Gur-Eyal Sela, Ionel Gog, Justin Wong +9
Efficient vision works maximize accuracy under a latency budget. These works evaluate accuracy offline, one image at a time. However, real-time vision applications like autonomous…
Quick Question: Interrupting Users for Microtasks with Reinforcement Learning
Bo-Jhang Ho, Bharathan Balaji, Mehmet Koseoglu +3
Human attention is a scarce resource in modern computing. A multitude of microtasks vie for user attention to crowdsource information, perform momentary assessments, personalize se…
ORL: Reinforcement Learning Benchmarks for Online Stochastic Optimization Problems
Bharathan Balaji, Jordan Bell-Masterson, Enes Bilgin +7
Reinforcement Learning (RL) has achieved state-of-the-art results in domains such as robotics and games. We build on this previous work by applying RL algorithms to a selection of…
DeepRacer: Educational Autonomous Racing Platform for Experimentation with Sim2Real Reinforcement Learning
Bharathan Balaji, Sunil Mallya, Sahika Genc +10
DeepRacer is a platform for end-to-end experimentation with RL and can be used to systematically investigate the key challenges in developing intelligent control systems. Using the…
ACES -- Automatic Configuration of Energy Harvesting Sensors with Reinforcement Learning
Francesco Fraternali, Bharathan Balaji, Yuvraj Agarwal +1
Internet of Things forms the backbone of modern building applications. Wireless sensors are being increasingly adopted for their flexibility and reduced cost of deployment. However…
Pible: Battery-Free Mote for Perpetual Indoor BLE Applications
Francesco Fraternali, Bharathan Balaji, Yuvraj Agarwal +2
Smart building applications require a large-scale deployment of sensors distributed across the environment. Recent innovations in smart environments are driven by wireless networke…
Scaling Configuration of Energy Harvesting Sensors with Reinforcement Learning
Francesco Fraternali, Bharathan Balaji, Rajesh Gupta
With the advent of the Internet of Things (IoT), an increasing number of energy harvesting methods are being used to supplement or supplant battery based sensors. Energy harvesting…
Did you hear that? Adversarial Examples Against Automatic Speech Recognition
Moustafa Alzantot, Bharathan Balaji, Mani Srivastava
Speech is a common and effective way of communication between humans, and modern consumer devices such as smartphones and home hubs are equipped with deep learning based accurate a…
EmbedInsight: Automated Grading of Embedded Systems Assignments
Hao Li, Bo-Jhang Ho, Bharathan Balaji +3
Grading in embedded systems courses typically requires a face-to-face appointment between the student and the instructor because of experimental setups that are only available in l…
Modeling Actuation Constraints for IoT Applications
Bharathan Balaji, Brad Campbell, Amit Levy +8
Internet of Things (IoT) promises to bring ease of monitoring, better efficiency and innovative services across many domains with connected devices around us. With information from…
Managing Commercial HVAC Systems: What do Building Operators Really Need?
Bharathan Balaji, Nadir Weibel, Yuvraj Agarwal
Buildings form an essential part of modern life; people spend a significant amount of their time in them, and they consume large amounts of energy. A variety of systems provide ser…
Quiver: Using Control Perturbations to Increase the Observability of Sensor Data in Smart Buildings
Jason Koh, Bharathan Balaji, Vahideh Akhlaghi +2
Modern buildings consist of hundreds of sensors and actuators for monitoring and operation of systems such as HVAC, light and security. To enable portable applications in next gene…
Genie: A Longitudinal Study Comparing Physical and Software-augmented Thermostats in Office Buildings
Bharathan Balaji, Jason Koh, Nadir Weibel +1
Thermostats are primary interfaces for occupants of office buildings to express their comfort preferences. However, standard thermostats are often ineffective due to inaccessibilit…
HVACMeter: Apportionment of HVAC Power to Thermal Zones and Air Handler Units
Jason Koh, Bharathan Balaji, Rajesh Gupta +1
Heating, Ventilation and Air Conditioning (HVAC) systems consume almost half of the total energy use of commercial buildings. To optimize HVAC energy usage, it is important to unde…