Conversational assistive robots have the potential to help human users accomplish a wide range of daily tasks, such as cooking meals, performing exercises, or operating …
Large pre-trained transformer-based language models (PTLMs) have recently dominated the state-of-the-art in Information Retrieval tasks such as web search and question answering. Despite the advantages …
I developed two public data systems deployed longitudinally in the wild: Drafty and Sketchy. Drafty is a tabular dataset of Computer Science professors that has …
With few exceptions, robots today are unable to quickly acquire new manipulation skills in the real world. The modern data-driven approach to skill learning is …
Combinatorial optimization problems are fundamental in many real-world applications, where the goal is to find the optimal or near-optimal solution to a problem subject to …
Neural networks continually surprise us. Many of these surprises are cases of impressive generalization. Others, however, are embarrassing failures. For example, large language models can …
Generally intelligent agents must learn and plan in complex environments, with sensors and actuators that support various behaviors and tasks. This complexity hinders decision making, …
The current dominant way to use neural networks for language applications is to pretrain a language model (LM): Given a large corpus, randomly mask out …
Humans are able to solve complex problems by distilling their knowledge of the world into simplified task-relevant representations and creating plans to achieve their goals. …
The adoption of cloud services is on the rise, with both personal and corporate data being stored on third-party server databases. In order to protect …
In the era of big data, distributed data stores have become an indispensable necessity. Almost all these data stores store sensitive data such as our …
Over the last decade, Python emerged as the language of choice for data processing and model building. Yet, the developer productivity of Python comes at …
This dissertation studies the canonical tasks of database compression and similarity search, and demonstrates how advanced deep learning models can be used to develop effective …
Indoor scenes and the shapes that comprise them play a central role in our daily experiences. As virtual 3D representations of these environments become increasingly …
At present, there are many blockchains, each competing with the others to gain market share and dominance. It is expected that multiple blockchains will continue …
Decentralized finance (or DeFi) has become a booming area of applied distributed computing. Over the course of the last several years, many mechanisms and distributed …
The digitization of the economy has had a revolutionary impact on society, and, today, people use digital web services to conduct essential daily activities, such …