Identifying the influence of training data for data cleansing can improve the accuracy of deep learning. An approach with stochastic gradient descent (SGD) called SGD-influence to calculate the influence scores was proposed, but, the calculation costs are expensive. It is necessary to temporally store the parameters of the model during training phase for inference phase to calculate influence sores. In close connection with the previous method, we propose a method to reduce cache files to store the parameters in training phase for calculating inference score. We only adopt the final parameters in last epoch for influence functions calculation. In our experiments on classification, the cache size of training using MNIST dataset with our approach is 1.236 MB. On the other hand, the previous method used cache size of 1.932 GB in last epoch. It means that cache size has been reduced to 1/1,563. We also observed the accuracy improvement by data cleansing with removal of negatively influential data using our approach as well as the previous method. Moreover, our simple and general proposed method to calculate influence scores is available on our auto ML tool without programing, Neural Network Console. The source code is also available.

Our tool provides an elegant user interface to design, train and evaluate neural network models. Deep learning technologies deserve to be used in practice more widely. This has been our anticipation since 2010, when we have started research and development involving deep learning, and continued to see its powers since then.

To further enhance its research capabilities Eco Marine Power announced today that it will begin using the Neural Network Console provided by Sony Network Communications Inc., as part of a strategy to incorporate Artificial Intelligence (AI) into various ongoing ship related technology projects including the further development of the patented Aquarius MRE (Marine Renewable Energy) and EnergySail. The Neural Network Console is an integrated development environment using deep learning for AI creation and has been used in deep learning applied technology development within Sony since 2015. Various functions are included such as recognition technology and a full-fledged GUI (graphical user interface) and these allow for deep learning programs to be developed. Deep learning refers to a form of machine learning that uses neural networks modelled after the human brain and is notable for its high versatility with applications in a wide variety of fields including signal processing, and robotics. An initial area of focus will be on studying how the Neural Network Console and AI can assist with the development of the automated control system for EMP's EnergySail.