UIUC Machine Learning Seminar (CS 591 MLR)

Welcome to the Machine Learning Seminar at the University of Illinois Urbana-Champaign!

  • The seminar is part of CS 591 MLR, whose faculty instructors are Arindam Banerjee and Han Zhao. To obtain a passing grade, students are required to participate in at least 75% of the talk. We will specify other logistics through mass mails.
  • For general audience, we have a mailing list that you can subscribe to.

Please find below the information of this semester (Fall 2024).

Physical Location: SC 3401 (except Sep 20 in SC 4124)

Zoom: https://illinois.zoom.us/j/85637981641?pwd=OJwoy8urE6RubYndg6ePK75Nxj5lN0.1

Time: Friday 2:00-3:30pm CT, Sep 6 – Dec 6

Mailing List:  uiuc-ml-reading-group@lists.illinois.edu (To Subscribe)

Student Organizers: Philip Amortila, Rohan Deb, and Yuzheng Hu

Dates
Location
Speaker
Affiliation
Title
09/06/2024Hybrid/Siebel 3401Wei Xiong
UIUC
Scalable Preference Learning for Large Language Model Post Training [slides]
09/13/2024Hybrid/Siebel 3401Mintong KangUIUCCertified Generation Risks for Retrieval-Augmented Language Models [slides]
09/20/2024RemoteKeegan HarrisCMUTowards Bridging Causal Inference and Algorithmic Decision-Making [slides]
09/27/2024RemoteRosie ZhaoHarvardEvaluating Optimizers for Language Model Training: From SGD to Adam (and Beyond?) [slides]
10/04/2024Hybrid/Siebel 3401Chi HanUIUCToward A Physiology of LM Representations [slides]
10/11/2024RemoteJingfeng WuUC BerkeleyReimaging gradient descent: Large stepsize oscillation and acceleration
10/18/2024RemoteJunjie HuUW MadisonAligning Foundation Models with the Open Multifaceted World
10/25/2024RemoteZak MhammediGoogle ResearchThe Power of Resets in Online Reinforcement Learning
11/01/2024RemoteDylan FosterMicrosoft ResearchIs Behavior Cloning All You Need? Understanding Horizon in Imitation Learning
11/08/2024RemoteHao WangRutgersBayesian Deep Learning: From Reliable Neural Networks to Interpretable Foundation Models
11/15/2024RemoteKaiwen WangCornellUnreasonable Effectiveness of Distributional Reinforcement Learning
11/22/2024RemoteHongyang ZhangUWaterlooEAGLE and EAGLE-2: Lossless Inference Acceleration for LLMs
11/29/2024N/AN/AFall Break
12/06/2024N/AN/AN/A

Sites of previous semesters:

Spring 2024

In loving memory of Nathanael Assefa, whose bright spirit was taken from us far too soon. We will forever miss him, both as a brilliant leader and a dear friend.