Neural Networks (NN), although successfully applied to several Artificial Intelligence tasks, are often unnecessarily over-parametrised. In edge/fog computing, this might make their training prohibitive on resource-constrained devices, contrasting with the current trend of decentralising intelligence from remote data centres to local constrained devices. Therefore, we investigate the problem of training effective NN models on constrained devices having a fixed, potentially small, memory budget. We target techniques that are both resource-efficient and performance effective while enabling significant network compression. Our Dynamic Hard Pruning (DynHP) technique incrementally prunes the network during training, identifying neurons that marginally contribute to the model accuracy. DynHP enables a tunable size reduction of the final neural network and reduces the NN memory occupancy during training. Freed memory is reused by a \emph{dynamic batch sizing} approach to counterbalance the accuracy degradation caused by the hard pruning strategy, improving its convergence and effectiveness. We assess the performance of DynHP through reproducible experiments on three public datasets, comparing them against reference competitors. Results show that DynHP compresses a NN up to $10$ times without significant performance drops (up to $3.5\%$ additional error w.r.t. the competitors), reducing up to $80\%$ the training memory occupancy.
翻译:神经网络(NN)虽然成功地应用于了几项人工智能任务,但往往不必要地过度偏差。在边缘/信息计算中,这可能会使其培训在资源限制装置上令人望而却步,这与目前将情报从远程数据中心分散到地方限制装置的趋势形成对照。因此,我们调查了对限制装置的有效NN模型进行培训的问题,这些模型具有固定的、潜在的小记忆预算。我们针对的是资源效率和性能都有效,同时又能促成重要的网络压缩的技术。我们在培训期间,我们动态硬质硬质普林金(DynHP)技术在对网络进行逐步完善,查明对模型准确性稍有帮助的神经元。DynHP使得最终神经网络的金枪鱼规模缩小,并在培训期间减少NNN的记忆占用率。自由记忆被一种固定、潜在小的批量化的方法重新利用,以抵消硬性裁剪裁战略造成的精确性退化,改进了它的趋同性和有效性。我们评估DynHP(DynHP)(DynHP)的绩效,方法是通过在三个公共数据集上进行再生实验,将它们与参考竞争者进行比较。结果显示DynHP(DynHP)将DynHP(DynHP)比低值为10美元;结果显示一个显著的硬性硬性硬性硬性硬性硬性硬性硬性运行到不至10次的硬性性能到高级性能,不至高级性硬性硬性能。结果。