Safe Active Learning for Time-Series Modeling with Gaussian Processes

Details

Location: Kulturaggregat | Hildastraße 5 · Freiburg im Breisgau

Our next speaker is Christoph Zimmer, he is a research scientist at the Bosch Center for Artificial Intelligence (BCAI). His research interests include calibration, parameter identifiability, safe active learning, and stochastic processes. Before working at Bosch, he obtained his Ph.D. from the Heidelberg University in interdisciplinary mathematics and he collaborated with systems biologists and epidemiologists during his postdocs at the center for modeling and simulations in the biosciences in Heidelberg and Yale School of Public Health.

He will talk about "Safe Active Learning for Time-Series Modeling with Gaussian Processes" and the BCAI. After the talk, there is time for an open discussion. We invite you to stay and have a get-together. This event is free and open for everyone who is interested in the broad field of AI.

Abstract of the talk:
Learning time-series models is useful for many applications, such as simulation and forecasting. In this study, we consider the problem of actively learning time-series models while taking given safety constraints into account. For time-series modeling we employ a Gaussian process with a nonlinear exogenous input structure. The proposed approach generates data appropriate for time series model learning, i.e. input and output trajectories, by dynamically exploring the input space. The approach parametrizes the input trajectory as consecutive trajectory sections, which are determined stepwise given safety requirements and past observations. We analyze the proposed algorithm and evaluate it empirically on a technical application. The results show the effectiveness of our approach in a realistic technical use case.

Further information: NeurIPS 2018 paper https://papers.nips.cc/paper/7538-safe-active-learning-for-time-series-modeling-with-gaussian-processes