Inbar Seroussi

17 papers A* 3Journal 14

Year	Rank	Type	Title / Venue / Authors
2025	J	`jnl`	A Geometric Unification of Generative AI with Manifold-Probabilistic Projection Models. CoRR Leah Bar, Liron Mor-Yosef, Shai Zucker, Neta Shoham, Inbar Seroussi, Nir A. Sochen
2025	J	`jnl`	Applications of Statistical Field Theory in Deep Learning. CoRR Zohar Ringel, Noa Rubin, Edo Mor, Moritz Helias, Inbar Seroussi
2025	J	`jnl`	Better Rates for Private Linear Regression in the Proportional Regime via Aggressive Clipping. CoRR Simone Bombari, Inbar Seroussi, Marco Mondelli
2025	J	`jnl`	Dimension-Free Minimax Rates for Learning Pairwise Interactions in Attention-Style Models. CoRR Shai Zucker, Xiong Wang, Fei Lu, Inbar Seroussi
2025	J	`jnl`	Exact Dynamics of Multi-class Stochastic Gradient Descent. CoRR Elizabeth Collins-Woodfin, Inbar Seroussi
2025	A*	`conf`	From Kernels to Features: A Multi-Scale Adaptive Theory of Feature Learning. ICML Noa Rubin, Kirsten Fischer, Javed Lindner, Inbar Seroussi, Zohar Ringel, Michael Krämer, Moritz Helias
2025	J	`jnl`	From Kernels to Features: A Multi-Scale Adaptive Theory of Feature Learning. CoRR Noa Rubin, Kirsten Fischer, Javed Lindner, David Dahmen, Inbar Seroussi, Zohar Ringel, Michael Krämer, Moritz Helias
2024	A*	`conf`	Grokking as a First Order Phase Transition in Two Layer Networks. ICLR Noa Rubin, Inbar Seroussi, Zohar Ringel
2024	J	`jnl`	Spectral-bias and kernel-task alignment in physically informed neural networks. Mach. Learn. Sci. Technol. Inbar Seroussi, Asaf Miron, Zohar Ringel
2024	A*	`conf`	The High Line: Exact Risk and Learning Rate Curves of Stochastic Adaptive Learning Rate Algorithms. NeurIPS Elizabeth Collins-Woodfin, Inbar Seroussi, Begoña García Malaxechebarría, Andrew W. Mackenzie, Elliot Paquette, Courtney Paquette
2023	J	`jnl`	Droplets of Good Representations: Grokking as a First Order Phase Transition in Two Layer Networks. CoRR Noa Rubin, Inbar Seroussi, Zohar Ringel
2023	J	`jnl`	Hitting the High-Dimensional Notes: An ODE for SGD learning dynamics on GLMs and multi-index models. CoRR Elizabeth Collins-Woodfin, Courtney Paquette, Elliot Paquette, Inbar Seroussi
2023	J	`jnl`	Spectral-Bias and Kernel-Task Alignment in Physically Informed Neural Networks. CoRR Inbar Seroussi, Asaf Miron, Zohar Ringel
2023	J	`jnl`	Speed Limits for Deep Learning. CoRR Inbar Seroussi, Alexander A. Alemi, Moritz Helias, Zohar Ringel
2022	J	`jnl`	Lower Bounds on the Generalization Error of Nonlinear Learning Models. IEEE Trans. Inf. Theory Inbar Seroussi, Ofer Zeitouni
2021	J	`jnl`	Lower Bounds on the Generalization Error of Nonlinear Learning Models. CoRR Inbar Seroussi, Ofer Zeitouni
2021	J	`jnl`	Separation of scales and a thermodynamic description of feature learning in some CNNs. CoRR Inbar Seroussi, Zohar Ringel