NeurIPS 2024 presentations from our group

Main Conference:

1. Partial observation can induce mechanistic mismatches in data-constrained models of neural dynamics
William Qian · Jacob Zavatone-Veth · Ben Ruben · Cengiz Pehlevan
East Exhibit Hall A-C #3807
Fri 13 Dec 1 p.m. CST — 4 p.m. CST

2. Infinite Limits of Multi-head Transformer Dynamics
Blake Bordelon · Hamza Chaudhry · Cengiz Pehlevan
East Exhibit Hall A-C #4804
Fri 13 Dec 1 p.m. CST — 4 p.m. CST

Workshops:

1. In-Context Learning by Linear Attention: Exact Asymptotics and Experiments
Yue Lu, Mary Letey, Jacob A Zavatone-Veth, Anindita Maiti, Cengiz Pehlevan
Mathematics of Modern Machine Learning (M3L)

2. Adversarially-robust representation learning through spectral regularization of features
Sheng Yang, Jacob A Zavatone-Veth, Cengiz Pehlevan
NeurReps

3. Partial observation can induce mechanistic mismatches in data-constrained RNNs
William Qian, Jacob A Zavatone-Veth, Benjamin Samuel Ruben, Cengiz Pehlevan
NeuroAI: Fusing Neuroscience and AI for Intelligent Solutions

4. A Hopfield network model of neuromodulatory arousal state,
Mo Osman, Kai Fox, Joshua Stern
NeuroAI: Fusing Neuroscience and AI for Intelligent Solutions