paper-with-me

Papers

HyperSparse Neural Networks: Shifting Exploration to Exploitation through Adaptive Regularization

2023-08-14 · Patrick Glandorf, Timo Kaiser, Bodo Rosenhahn

Sparse neural networks are a key factor in developing resource-efficient machine learning applications. We propose the novel and powerful sparse learning method Adaptive Regularized Training (ART) to compress dense into sparse networks. Instead of the commonly used binary mask during training to reduce the number of model weights, we inherently shrink weights close to zero in an iterative manner with increasing weight regularization. Our method compresses the pre-trained model knowledge into the weights of highest magnitude. Therefore, we introduce a novel regularization loss named HyperSparse that exploits the highest weights while conserving the ability of weight exploration. Extensive experiments on CIFAR and TinyImageNet show that our method leads to notable performance gains compared to other sparsification methods, especially in extremely high sparsity regimes up to 99.8 percent model sparsity. Additional investigations provide new insights into the patterns that are encoded in weights with high magnitudes.

📄 PDF Abstract BibTeX arXiv:2308.07163

Code (1)

greenautoml4fas/hypersparse 공식 구현 pytorch

Tasks

Sparse Learning

Similar Papers 제목 키워드 기반

Reward Shifting for Optimistic Exploration and Conservative Exploitation

2021-09-29 · Hao Sun, Lei Han, Jian Guo, Bolei Zhou

In this work, we study the simple yet universally applicable case of reward shaping, the linear transformation, in value-based Deep Reinforcement Learning. We show that reward shifting, as the simplest linear reward tran…

continuous-controlContinuous ControlDeep Reinforcement LearningOffline RL

Optimistic Curiosity Exploration and Conservative Exploitation with Linear Reward Shaping

2022-09-15 · Hao Sun, Lei Han, Rui Yang, Xiaoteng Ma 외

In this work, we study the simple yet universally applicable case of reward shaping in value-based Deep Reinforcement Learning (DRL). We show that reward shifting in the form of the linear transformation is equivalent to…

continuous-controlContinuous ControlDeep Reinforcement LearningOffline RL

The Exploration-Exploitation Dilemma Revisited: An Entropy Perspective

2024-08-19 · Renye Yan, Yaozhong Gan, You Wu, Ling Liang 외

The imbalance of exploration and exploitation has long been a significant challenge in reinforcement learning. In policy optimization, excessive reliance on exploration reduces learning efficiency, while over-dependence …

MuJoCo

Opt3DGS: Optimizing 3D Gaussian Splatting with Adaptive Exploration and Curvature-Aware Exploitation

2025-11-17 · Ziyang Huang, Jiagang Chen, Jin Liu, Shunping Ji arxiv

3D Gaussian Splatting (3DGS) has emerged as a leading framework for novel view synthesis, yet its core optimization challenges remain underexplored. We identify two key issues in 3DGS optimization: entrapment in suboptim…

Novel View Synthesis

Learn the Ropes, Then Trust the Wins: Self-imitation with Progressive Exploration for Agentic Reinforcement Learning

2025-09-26 · Yulei Qin, Xiaoyu Tan, Zhengbao He, Gang Li 외 arxiv

Reinforcement learning (RL) is the dominant paradigm for sharpening strategic tool use capabilities of LLMs on long-horizon, sparsely-rewarded agent tasks, yet it faces a fundamental challenge of exploration-exploitation…

Reinforcement Learning