We propose a novel approach for loss reserving based on deep neural networks. The approach allows for joint modeling of paid losses and claims outstanding, and incorporation of heterogeneous inputs. We validate the models on loss reserving data across lines of business, and show that they improve on the predictive accuracy of existing stochastic methods. The models require minimal feature engineering and expert input, and can be automated to produce forecasts more frequently than manual workflows.
Machine learning methods have garnered increasing interest among actuaries in recent years. However, their adoption by practitioners has been limited, partly due to the lack of transparency of these methods, as compared to generalized linear models. In this paper, we discuss the need for model interpretability in property & casualty insurance ratemaking, propose a framework for explaining models, and present a case study to illustrate the framework.
1 January 2020 in The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec)
This study proposed an effective imitation learning humanoid robot platform based on a Game Engine which considered usual creators of 3DCG animator or game creator’s usual development environment. We verify the proposed platform with an actual imitation learning task which is trained our robot to learn to generate 10 different action patterns. Each action pattern contains time-series motor angle information, facial animation command and voice command. Finally, we evaluate the man-hour cost through the instructor of the Japanese Industrial Standards(JIS Z 8141-1227) and show a 60% reduction of time cost for executing the same manner to a similar setup.