Skip to content
conftrace_

information exponent

7 papers

Explore in graph

Co-occurring keywords · papers mentioning both

single-index model (5) stochastic gradient descent (3) two-layer neural network (2) gradient descent (2) sample complexity (2) representation learning (2) information-theoretic limit (1) polynomial learning (1) ridge function (1) gradient descent training (1)

Papers

Mixture of Experts Provably Detect and Learn the Latent Cluster Structure in Gradient-Based Learning ICML 2025 Learning sum of diverse features: computational hardness and efficient gradient-based training for ridge combinations COLT 2024 Neural network learns low-dimensional polynomials with SGD near the information-theoretic limit NeurIPS 2024 Online Learning and Information Exponents: The Importance of Batch size & Time/Complexity Tradeoffs ICML 2024 Symmetric Single Index Learning ICLR 2024 The Benefits of Reusing Batches for Gradient Descent in Two-Layer Networks: Breaking the Curse of Information and Leap Exponents ICML 2024 On Single-Index Models beyond Gaussian Data NeurIPS 2023