Boris Hanin

Publication Activity (10 Years)

Years Active: 2017-2024
Publications (10 Years): 39

Top Topics

Top Venues

Publications

Lingjiao Chen, Jared Quincy Davis, Boris Hanin, Peter Bailis, Ion Stoica, Matei Zaharia, James Zou
Are More LLM Calls All You Need? Towards Scaling Laws of Compound Inference Systems. CoRR (2024)
Wuyang Chen, Junru Wu, Zhangyang Wang, Boris Hanin
Principled Architecture-aware Scaling of Hyperparameters. CoRR (2024)
Blake Bordelon, Lorenzo Noci, Mufan Bill Li, Boris Hanin, Cengiz Pehlevan
Depthwise Hyperparameter Transfer in Residual Networks: Dynamics and Scaling Limit. ICLR (2024)
Wuyang Chen, Junru Wu, Zhangyang Wang, Boris Hanin
Principled Architecture-aware Scaling of Hyperparameters. ICLR (2024)
Jared Quincy Davis, Boris Hanin, Lingjiao Chen, Peter Bailis, Ion Stoica, Matei Zaharia
Networks of Networks: Complexity Class Principles Applied to Compound AI Systems Design. CoRR (2024)
Boris Hanin, Alexander Zlokapa
Bayesian Inference with Deep Weakly Nonlinear Networks. CoRR (2024)
Blake Bordelon, Lorenzo Noci, Mufan Bill Li, Boris Hanin, Cengiz Pehlevan
Depthwise Hyperparameter Transfer in Residual Networks: Dynamics and Scaling Limit. CoRR (2023)
Gage DeZoort, Boris Hanin
Principles for Initialization and Architecture Selection in Graph Neural Networks with ReLU Activations. CoRR (2023)
Stefano Favaro, Boris Hanin, Domenico Marinucci, Ivan Nourdin, Giovanni Peccati
Quantitative CLTs in Deep Neural Networks. CoRR (2023)
Samy Jelassi, Boris Hanin, Ziwei Ji, Sashank J. Reddi, Srinadh Bhojanapalli, Sanjiv Kumar
Depth Dependence of μP Learning Rates in ReLU MLPs. CoRR (2023)
Gaurav Iyer, Boris Hanin, David Rolnick
Maximal Initial Learning Rates in Deep ReLU Networks. ICML (2023)
Yasaman Bahri, Boris Hanin, Antonin Brossollet, Vittorio Erba, Christian Keup, Rosalba Pacelli, James B. Simon
Les Houches Lectures on Deep Learning at Large & Infinite Width. CoRR (2023)
Wuyang Chen, Wei Huang, Xinyu Gong, Boris Hanin, Zhangyang Wang
Deep Architecture Connectivity Matters for Its Convergence: A Fine-Grained Analysis. CoRR (2022)
Boris Hanin, Alexander Zlokapa
Bayesian Interpolation with Deep Linear Networks. CoRR (2022)
Boris Hanin
Correlation Functions in Random Fully Connected Neural Networks at Finite Width. CoRR (2022)
Gaurav Iyer, Boris Hanin, David Rolnick
Maximal Initial Learning Rates in Deep ReLU Networks. CoRR (2022)
Boris Hanin, Ryan S. Jeong, David Rolnick
Deep ReLU Networks Preserve Expected Length. ICLR (2022)
Wuyang Chen, Wei Huang, Xinyu Gong, Boris Hanin, Zhangyang Wang
Deep Architecture Connectivity Matters for Its Convergence: A Fine-Grained Analysis. NeurIPS (2022)
Daniel A. Roberts, Sho Yaida, Boris Hanin
The Principles of Deep Learning Theory. CoRR (2021)
Boris Hanin
Ridgeless Interpolation with Shallow ReLU Networks in $1D$ is Nearest Neighbor Curvature Extrapolation and Provably Generalizes on Lipschitz Functions. CoRR (2021)
Boris Hanin, Ryan Jeong, David Rolnick
Deep ReLU Networks Preserve Expected Length. CoRR (2021)
Ronald A. DeVore, Boris Hanin, Guergana Petrova
Neural network approximation. Acta Numer. 30 (2021)
Boris Hanin, Yi Sun
How Data Augmentation affects Optimization for Linear Regression. NeurIPS (2021)
Boris Hanin
Random Neural Networks in the Infinite Width Limit as Gaussian Processes. CoRR (2021)
Ronald A. DeVore, Boris Hanin, Guergana Petrova
Neural Network Approximation. CoRR (2020)
Boris Hanin, Yi Sun
Data augmentation as stochastic optimization. CoRR (2020)
Boris Hanin, Mihai Nica
Finite Depth and Width Corrections to the Neural Tangent Kernel. ICLR (2020)
Boris Hanin, David Rolnick
Deep ReLU Networks Have Surprisingly Few Activation Patterns. CoRR (2019)
Boris Hanin, David Rolnick
Complexity of Linear Regions in Deep Networks. CoRR (2019)
Boris Hanin, David Rolnick
Deep ReLU Networks Have Surprisingly Few Activation Patterns. NeurIPS (2019)
Boris Hanin, David Rolnick
Complexity of Linear Regions in Deep Networks. ICML (2019)
Ingrid Daubechies, Ronald A. DeVore, Simon Foucart, Boris Hanin, Guergana Petrova
Nonlinear Approximation and (Deep) ReLU Networks. CoRR (2019)
Boris Hanin, Mihai Nica
Finite Depth and Width Corrections to the Neural Tangent Kernel. CoRR (2019)
Boris Hanin, David Rolnick
How to Start Training: The Effect of Initialization and Architecture. NeurIPS (2018)
Boris Hanin
Which Neural Net Architectures Give Rise To Exploding and Vanishing Gradients? CoRR (2018)
Boris Hanin, David Rolnick
How to Start Training: The Effect of Initialization and Architecture. CoRR (2018)
Boris Hanin
Which Neural Net Architectures Give Rise to Exploding and Vanishing Gradients? NeurIPS (2018)
Boris Hanin
Universal Function Approximation by Deep Neural Nets with Bounded Width and ReLU Activations. CoRR (2017)
Boris Hanin, Mark Sellke
Approximating Continuous Functions by ReLU Nets of Minimal Width. CoRR (2017)