Welcome to the homepage of the Machine Learning Research Group at Tufts University!
Jan 2020: Launch of the T-TRIPODS Institute across Tufts
We are launching a new campus-wide initiative: The Tufts Transdisciplinary Research In Principles of Data Science Institute (T-TRIPODS). This will bring together interdisciplinary research across departments and schools at Tufts to advance the understanding of foundations of data science, with support from the National Science Foundation (NSF). The focus is on fostering collaboration between core methodologists in the CS, ECE, and Math departments on three targeted methodological areas: graph and tensor representations, spatiotemporal modeling, and providing assurances of quality, fairness, privacy and trust. The institute will also push those advances to impact key application areas across the university.
More about the T-TRIPODS institute at Tufts:
Dec 2019: Several papers from Tufts accepted at AAAI 2020
Prof. Liping Liu, students Gabriel Appleby and Linfeng Liu, and Prof. Avery Cohn wrote "Kriging Convolutional Networks"
Prof. Mike Hughes and several external colleagues have a paper Regional Tree Regularization for Interpretability in Deep Neural Networks
Nov 2019: Team research project on Open-World AI is funded by DARPA via the SAIL-ON program
This project is led by PI Matthias Scheutz, with co-PIs Jivko Sinapov, Mike Hughes, and Liping Liu. SAIL-ON stands for "Science of Artificial Intelligence and Learning for Open-world Novelty". More about the program: DARPA SAIL-ON Announcement.
Sep 2019: Prof. Liping Liu and Prof. Mike Hughes' project on inference for large graphical models is funded by NSF
The project, titled by "Amortized Inference for Large-Scale Graphical Models", wins an NSF Small grant from the Division of Information and Intelligent Systems. Here is the link.
Mar 2019: Prof. Liping Liu's project on self-attention neural networks is funded by NSF
The project, titled by "CRII: RI: Self-Attention through the Bayesian Lens", wins an NSF CRII grant. Here is the link.
Feb 2019: Linfeng Liu's paper on amortized inference for Gaussian processes is published at AISTATS 2019
The paper "Amortized Variational Inference with Graph Convolutional Networks for Gaussian Processes" [link] proposes a new efficient inference method for Gaussian processes. It decomposes the inference task into sub-tasks at local neighborhoods and use a shared inference network to amortize computation cost.
Dec 2018: Tufts Prof. Mike Hughes is organizing two workshops at the upcoming NeurIPS 2018 conference
Aug. 2018: Welcome Prof. Mike Hughes
Mike joins us as new faculty in the Computer Science dept.Meet Mike
Aug. 2017: Welcome Prof. Liping Liu
Liping joins us as new faculty in the Computer Science dept.Meet Liping