Now showing items 11-20 of 219
Multi-task Hierarchical Reinforcement Learning for Compositional Tasks
(2021)
This thesis presents the algorithms for solve multiple compositional tasks with high sample efficiency and strong generalization ability.
Central to this work is the subtask graph which models the structure in compositional ...
Social Reference Processing with Collaborative Human-AI Systems
(2023)
In this dissertation, I introduce several human-AI systems that can understand and process social references — language that invokes connotations by overlaying parts of social and cultural contexts. I argue that developing ...
Long-Horizon Planning Under Uncertainty and Geometric Constraints for Mobile Manipulation by Autonomous Humanoid Robots
(2023)
Autonomous humanoid robots have the potential to perform critical and labor-intensive tasks that could go a long way to improve upon the quality of human life. To realize this potential, an autonomous humanoid robot must ...
In-SRAM Computing for Neural Network Acceleration
(2022)
For decades, the computing paradigm has been composed of separate memory and compute units. Processing-in-Memory(PIM) has often been proposed as a solution to break past the memory wall. With PIM, compute logic is moved ...
Architectural Enhancements for Data Transport in Datacenter Systems
(2021)
Datacenter systems run myriad applications, which frequently communicate with each other and/or Input/Output (I/O) devices—including network adapters, storage devices, and accelerators. Due to the growing speed of I/O ...
Automating the Detection and Correction of Failures in Modern Persistent Memory Systems
(2023)
Modern software systems are deeply embedded into our daily lives; the failures of these systems can therefore result in massive real-world harm. Consequently, considerable resources are spent finding and fixing bugs in ...
Neural Language Generation for Content Adaptation: Explainable, Efficient Low-Resource Text Simplification and Evaluation
(2023)
There are rich opportunities to reduce the language complexity of professional content (either human-written or computer-generated) and make it accessible to a broad audience. As a sub-task of natural language generation ...
Processing Image Data from Unstructured Environments
(2023)
Advanced mobility research centers capture large amounts of data from ground vehicle systems during development and experimentation in both manned and autonomous operations. This exponential growth of digital image data ...
Hazy Oracles in Deep Learning
(2023)
While deep learning problems are often motivated as enabling technologies for human-computer interaction---a robot, for example, must align natural language referents and sensor readings to operate in a human world---assumptions ...
Label Uncertainty and Learning Using Partially Available Privileged Information for Clinical Decision Support: Applications in Detection of Acute Respiratory Distress Syndrome
(2021)
Artificial intelligence and machine learning have the potential to transform health care by deriving new and important insights from the vast amount of data generated during routine delivery of healthcare. The digitization ...