Dissertation Defenses

Enforcing Realism and Temporal Consistency for Large-Scale Video Inpainting

Cautiously Optimistic Program Analyses for Secure and Reliable Software

Efficient Resource Management for Deep Learning Clusters

Architectural Enhancements for Data Transport in Datacenter Systems

Towards Closing the Programmability-Efficiency Gap using Software-Defined Hardware

Towards Learning the Foundations of Manipulation Actions with Unguided Exploration

Enabling Hyperscale Web Services

Understanding User Cognition: from Spatial Ability to Code Writing and Review

Ride-Sharing Optimization Algorithms for Urban Commuting

A Comprehensive Computational Model of PRIMs Theory for Task-Independent Procedural Learning

Multi-task Hierarchical Reinforcement Learning for Compositional Tasks

Developing Trustworthy Hardware with Security-Driven Design & Verification

Sample efficient learning and generalization with text representations

On Learning Neural End-to-End Goal-Oriented Dialog Systems

Datacenter Architectures for the Microservices Era

Evaluating and Improving Internet Load Balancing with Large-Scale Latency Measurements

Improving the Generalization Performance of Deep Learning Models Applied to Clinical Time Series by Addressing Task Specific Structure

Expanding Task Diversity in Explanation-Based Interactive Task Learning

Computational Modeling and Design of Financial Markets: Towards Manipulation-Resistant and Expressive Markets

Streaming Architectures for Medical Image Reconstruction

A Holistic Solution for Reliability of 3D Parallel Systems

Leveraging Longitudinal Data for Personalized Prediction and Word Representations

Software Model Checking with Uninterpreted Functions

Design Optimization of In-Vehicle Network

Algebraic Frameworks for Cryptographic Primitives

Automatic Quantification and Prediction of Affect in Spoken Interactions

Models and Algorithms for Human Action Understanding

Customized Systems of Deep Neural Networks for Energy Efficiency