- Congrats to Christopher, Kevin, Yang, Juan, and Manzil on our AAAI 2020 paper: Exchangeable Generative Models with Flow Scans!
- Congrats to Yang, and Tianxiang on our AAAI 2020 paper: A Forest from the Trees: Generation through Neighborhoods!
- Congrats to Yifeng, and Marc on our AAAI 2020 paper: Deep Message Passing on Sets!
- Congrats to Eunbyung on our NeurIPS 2019 paper: Meta-Curvature!
Welcome to the LUPA Lab, a research group devoted to machine learning and artificial intelligence. We are looking to see what makes data tick, and understanding data at an aggregate, holistic level. The LUPA Lab is using techniques ranging from modern deep learning architectures to nonparametric statistics to make strides in areas like: high-dimensional density estimation and modeling; sequential modeling and RNNs; and learning over complex or structured data. Take a glance below for a word-cloud of our interests, or see our projects page for what we are working on.
We are exploring collective approaches (that exploit collections like sets and distributions) to bridge the gap between machine and human learning by providing further context than myopic point estimation approaches. We prefer simple estimators that make few assumptions; that is, flexible and powerful methods that are able to generalize and extrapolate. Further, we are developing techniques for analyzing massive datasets, both in terms of instances and covariates.
This work will help us solve problems like predicting whether a Twitter topic will go viral, or predicting the risk of disease given a person's functional brain data, or predicting the future distribution of dark matter particles. Application areas include: health and medicine; science; business; earth and climate; and computer vision.
Machine learning, artificial intelligence, nonparametric statistics, deep learning, statistical data mining, signal processing, graphical models, generative models, kernel methods, scalability, complex datasets, optimization, density estimation.