Jointly Training and Pruning CNNs via Learnable Agent Guidance and Alignment

arXiv:2403.19490v1 Announce Type: new
Abstract: Structural model pruning is a prominent approach used for reducing the computational cost of Convolutional Neural Networks (CNNs) before their deployment on resource-constrained devices. Yet, the majority of proposed ideas require a pretrained model before pruning, which is costly to secure. In this paper, we propose a novel structural pruning approach to jointly learn the weights and structurally prune architectures of CNN models. The core element of our method is a Reinforcement Learning (RL) agent …

abstract agent alignment arxiv cnns computational convolutional neural networks cost cs.cv deployment devices guidance ideas networks neural networks paper pruning training type via