Orchard View Room – Page 9 – SILO

SILO: Looped Transformers as Programmable Computers

Dimitris Tsipras, Stanford

SILO: Controlling the Structure of Inference and Learning in Neural Networks

Yisong Yue, CalTech

SILO: Dyamic network models: one has to do the math

Shankar Bhamidi, UNC

SILO: Benign Overfitting in Two-layer ReLU Networks

Quanquan Gu, UCLA

SILO: Bottom-up manifold learning with low distortion

Gal Mishne, UCSD

SILO: Bagging provides assumption-free stability

Jake Soloff, University of Chicago

SILO: Online Safe Learning and Decision Making

Lei Ying, University of Michigan

SILO: An invitation to spectral methods

Joshua Cape, University of Wisconsin–Madison

SILO: Faster Neural Network Training, Algorithmically

Jonathan Frankle, Mosaic/Harvard

SILO: Score-based Generative Modeling Secretly Minimizes the Wasserstein Distance

Kangwook Lee, University of Wisconsin–Madison