Synaptic pruning with MAP-elites
Synaptic pruning with MAP-elites
复制标题
DOI：
10.1145/3520304.3528813
复制
发表时间：
2022-07
期刊：
影响因子：
--
通讯作者：
Federico Da Rold;Olaf Witkovski;N. Aubert-Kato
中科院分区：
文献类型：
--
作者：
Federico Da Rold;Olaf Witkovski;N. Aubert-Kato
减少深度学习模型中的参数数量是机器学习当前面临的挑战。我们利用MAP-Elites在强化学习问题中照亮搜索空间的能力，面对具有不同数量连接的神经网络。在这项工作中，我们特别关注Open AI BipedalWalker [2]，这是一种广泛使用的强化学习基准，以及成功解决该任务的深度学习模型。由此产生的架构显示突触连接性减少了约90%，相当于监督学习或生成学习中通常采用的最先进的修剪技术。具体来说，我们的方法使我们能够评估稀疏性和辍学的各自影响。结果表明，在我们的泛化测试中，隐单元的稀疏性和丢失率与性能无关。
Reducing the number of parameters in deep learning models is a current challenge in machine learning. We exploit the capability of MAP-Elites of illuminating the search space in a reinforcement learning problem, confronting neural networks with a different number of connections. In this work, we focus specifically on the Open AI BipedalWalker [2], a widely employed reinforcement learning benchmark, and on a deep learning model that successfully solved that task. The resulting architectures show a reduction of the synaptic connectivity of approximately 90%, equivalent to the state-of-the-art pruning techniques usually employed in supervised or generative learning. Specifically, our approach allows us to evaluate the respective impacts of sparsity and dropout. Results show that both sparsity and dropouts of hidden units are uncorrelated to the performance in our generalization test.