paper-with-me

홈 › Papers

Revisiting Parameter Sharing for Automatic Neural Channel Number Search

2020-12-01 · NeurIPS 2020 12 · Jiaxing Wang, Haoli Bai, Jiaxiang Wu, Xupeng Shi, Junzhou Huang, Irwin King, Michael Lyu, Jian Cheng

Recent advances in neural architecture search inspire many channel number search algorithms~(CNS) for convolutional neural networks. To improve searching efficiency, parameter sharing is widely applied, which reuses parameters among different channel configurations. Nevertheless, it is unclear how parameter sharing affects the searching process. In this paper, we aim at providing a better understanding and exploitation of parameter sharing for CNS. Specifically, we propose affine parameter sharing~(APS) as a general formulation to unify and quantitatively analyze existing channel search algorithms. It is found that with parameter sharing, weight updates of one architecture can simultaneously benefit other candidates. However, it also results in less confidence in choosing good architectures. We thus propose a new strategy of parameter sharing towards a better balance between training efficiency and architecture discrimination. Extensive analysis and experiments demonstrate the superiority of the proposed strategy in channel configuration against many state-of-the-art counterparts on benchmark datasets.

📄 PDF Abstract BibTeX

Code (1)

haolibai/APS-channel-search 공식 구현 pytorch

Tasks

Neural Architecture Search

Similar Papers 제목 키워드 기반

Scaling Multi-Agent Reinforcement Learning with Selective Parameter Sharing

2021-02-15 · Filippos Christianos, Georgios Papoudakis, Arrasy Rahman, Stefano V. Albrecht

Sharing parameters in multi-agent deep reinforcement learning has played an essential role in allowing algorithms to scale to a large number of agents. Parameter sharing between agents significantly decreases the number …

Deep Reinforcement LearningMulti-agent Reinforcement Learningreinforcement-learningReinforcement Learning+1

Drastically Reducing the Number of Trainable Parameters in Deep CNNs by Inter-layer Kernel-sharing

2022-10-23 · Alireza Azadbakht, Saeed Reza Kheradpisheh, Ismail Khalfaoui-Hassani, Timothée Masquelier

Deep convolutional neural networks (DCNNs) have become the state-of-the-art (SOTA) approach for many computer vision tasks: image classification, object detection, semantic segmentation, etc. However, most SOTA networks …

Edge-computingimage-classificationImage Classificationobject-detection+2

Selectively Sharing Experiences Improves Multi-Agent Reinforcement Learning

2023-11-01 · NeurIPS 2023 11 · Matthias Gerstgrasser, Tom Danino, Sarah Keren

We present a novel multi-agent RL approach, Selective Multi-Agent Prioritized Experience Relay, in which agents share with other agents a limited number of transitions they observe during training. The intuition behind t…

Multi-agent Reinforcement Learningreinforcement-learningReinforcement Learning

Revisiting Dynamic Convolution via Matrix Decomposition

2021-03-15 · ICLR 2021 1 · Yunsheng Li, Yinpeng Chen, Xiyang Dai, Mengchen Liu 외

Recent research in dynamic convolution shows substantial performance boost for efficient CNNs, due to the adaptive aggregation of K static convolution kernels. It has two limitations: (a) it increases the number of convo…

Dimensionality Reduction

RecNets: Channel-wise Recurrent Convolutional Neural Networks

2019-05-28 · George Retsinas, Athena Elafrou, Georgios Goumas, Petros Maragos

In this paper, we introduce Channel-wise recurrent convolutional neural networks (RecNets), a family of novel, compact neural network architectures for computer vision tasks inspired by recurrent neural networks (RNNs). …

General Classificationimage-classificationImage Classification