Exploring ML-based compression for rapid release of astronomical alerts
The Rubin Observatory will see millions of "transients" a night, but cannot release all of these to the wider astronomical community without heavily filtering them. We will explore recent advances in machine learning techniques that produce optimal compressions to address this problem.
Efficient AI-driven System Optimizers with Adaptive Transformers
Modern AI-driven system optimizers are hitting a wall, crippled by massive computational costs and an inability to navigate the overwhelming complexity of high-dimensional data. This project tackles the array of bottlenecks that prevent these systems from scaling.
Building Better Agentic AI browsers
We aim to build the next generation of agentic AI browsers -- systems like OpenAI Atlas or Perplexity Comet -- with much faster performance while maintaining high accuracy.
Machine learning for cancer research
Interested in applying machine/deep learning for cancer research? We are looking for a part time computational research assistant/programmer to join our lab for an hourly paid position. Ideal for MS students or BS CIS/engineering students in their final year, who want to use their comp skills to make discoveries.
Computational Social Science
I am broadly interested in topics at the intersection of social science and computer science, with a particular interest in computational studies of media, collective intelligence, network diffusion, and common sense.
Computer Vision and Machine Learning Applied in Robotics Contexts
The central focus of our research program is versatile robotic control through tight integration with visual perception and other general-purpose high-bandwidth sensing modalities such as touch. Through this, our aim is to enable robots to operate usefully in our homes, workplaces, and everyday lives.
Cross-Cultural Language Analysis of Social Data
Applications of natural language to health and psychology, especially but not limited to – social norms, vaccine acceptance, and global mental health.
Learning Directed Operating System
We are building the next-generation Machine Learning-based Operating System to drive computing infrastructure toward high efficiency and performance.
Programming Languages, Software Verification
Learn about functional programming, low-level code, and compilers; build reliable software using interactive theorem proving.
Developing Biochemical & Computational Diagnostics in Mitochondrial Disease
We use human cell lines as well as statistical learning approaches to study biochemical abnormalities that occur in primary mitochondrial disease, with a core goal of improving our ability to rapidly diagnose mitochondrial disease.