← Authors

James Martens

16 papers · 2010–2024 · 5 conferences · across top CS/AI conferences

Achievements

🐣 Hot Topic Early Bird 🌈 Renaissance Researcher (5) πŸŒ‰ Interdisciplinary Bridge πŸ—ΊοΈ Taxonomy Completionist (10) 🧭 Keyword Pioneer 🐝 Cross-Pollinator (14) 🌍 Conference Polyglot (5) πŸƒ Academic Marathon (14) πŸ† Keyword Champion (2) πŸ’Ž Century Club (16) πŸš€ Conference Pioneer ❓ The Questioner

Conferences

NIPS (5) ICLR (4) ICML (4) JMLR (2) AISTATS (1)

Top co-authors

Guodong Zhang (4) Yee Whye Teh (2) Thore Graepel (2) Razvan Pascanu (2) Roger B Grosse (2) Sébastien Racanière (2) Roger Grosse (2) George Dahl (2) Ilya Sutskever (2) Aleksandar Botev (2)

Keywords

natural gradient (4) second-order optimization (3) convergence analysis (2) neural network (2) stochastic gradient descent (2) gradient descent (2) generative adversarial network (2) fisher information matrix (2) symplectic gradient adjustment (2) hamiltonian game (2) differentiable game (2) restricted boltzmann machine (2) kronecker factorization (2) nash equilibrium (1) probabilistic inference (1) markov chain monte carlo (1) neural network theory (1) adversarial robustness (1) representation learning (1) neural network optimization (1)

Papers

Normalization and effective learning rates in reinforcement learning NIPS 2024
Pre-training via Denoising for Molecular Property Prediction ICLR 2023
Deep Transformers without Shortcuts: Modifying Self-attention for Faithful Signal Propagation ICLR 2023
Deep Learning without Shortcuts: Shaping the Kernel with Tailored Rectifiers ICLR 2022
New Insights and Perspectives on the Natural Gradient Method JMLR 2020
Adversarial Robustness through Local Linearization NIPS 2019
Differentiable Game Mechanics JMLR 2019
Which Algorithmic Choices Matter at Which Batch Sizes? Insights From a Noisy Quadratic Model NIPS 2019
Fast Convergence of Natural Gradient Descent for Over-Parameterized Neural Networks NIPS 2019
The Mechanics of n-Player Differentiable Games ICML 2018
Kronecker-factored Curvature Approximations for Recurrent Neural Networks ICLR 2018
A Kronecker-factored approximate Fisher matrix for convolution layers ICML 2016
Optimizing Neural Networks with Kronecker-factored Approximate Curvature ICML 2015
On the Representational Efficiency of Restricted Boltzmann Machines NIPS 2013
On the importance of initialization and momentum in deep learning ICML 2013
Parallelizable Sampling of Markov Random Fields AISTATS 2010