conftrace_
← Optimization & Theory
Machine Learning › Optimization & Theory ›

Neural Network Optimization

3,649 papers

Papers per year

2
5
40
414
386
'05
'10
'15
'20
'25

Papers

Improving Deep Learning Optimization through Constrained Parameter Regularization NIPS 2024
A generalized neural tangent kernel for surrogate gradient learning NIPS 2024
LoFiT: Localized Fine-tuning on LLM Representations NIPS 2024
The Road Less Scheduled NIPS 2024
Stacking Your Transformers: A Closer Look at Model Growth for Efficient LLM Pre-Training NIPS 2024
On Convergence of Adam for Stochastic Optimization under Relaxed Assumptions NIPS 2024
Non-asymptotic Analysis of Biased Adaptive Stochastic Approximation NIPS 2024
Fundamental Convergence Analysis of Sharpness-Aware Minimization NIPS 2024
Pretraining with Random Noise for Fast and Robust Learning without Weight Transport NIPS 2024
Initialization is Critical to Whether Transformers Fit Composite Functions by Reasoning or Memorizing NIPS 2024
Data-Efficient Learning with Neural Programs NIPS 2024
ST$_k$: A Scalable Module for Solving Top-k Problems NIPS 2024
Ordered Momentum for Asynchronous SGD NIPS 2024
Real-Time Recurrent Learning using Trace Units in Reinforcement Learning NIPS 2024
Efficient Reinforcement Learning by Discovering Neural Pathways NIPS 2024
Evaluating the design space of diffusion-based generative models NIPS 2024
Beyond Slow Signs in High-fidelity Model Extraction NIPS 2024
DropBP: Accelerating Fine-Tuning of Large Language Models by Dropping Backward Propagation NIPS 2024
Predictor-Corrector Enhanced Transformers with Exponential Moving Average Coefficient Learning NIPS 2024
Nesterov acceleration despite very noisy gradients NIPS 2024
Implicit Optimization Bias of Next-token Prediction in Linear Models NIPS 2024
Why Do We Need Weight Decay in Modern Deep Learning? NIPS 2024
Identifying Equivalent Training Dynamics NIPS 2024
The Implicit Bias of Adam on Separable Data NIPS 2024
Take A Shortcut Back: Mitigating the Gradient Vanishing for Training Spiking Neural Networks NIPS 2024