Christoph Lüth

Daniel Harnack, Christoph Lüth, Lukas Gross et al.

Generating physical movement behaviours from their symbolic description is a long-standing challenge in artificial intelligence (AI) and robotics, requiring insights into numerical optimization methods as well as into formalizations from symbolic AI and reasoning. In this paper, a novel approach to finding a reward function from a symbolic description is proposed. The intended system behaviour is modelled as a hybrid automaton, which reduces the system state space to allow more efficient reinforcement learning. The approach is applied to bipedal walking, by modelling the walking robot as a hybrid automaton over state space orthants, and used with the compass walker to derive a reward that incentivizes following the hybrid automaton cycle. As a result, training times of reinforcement learning controllers are reduced while final walking speed is increased. The approach can serve as a blueprint how to generate reward functions from symbolic AI and reasoning.

Cezary Kaliszyk, Christoph Lüth

This EPTCS volume collects the post-proceedings of the 10th International Workshop On User Interfaces for Theorem Provers (UITP 2012), held as part of the Conferences on Intelligent Computer Mathematics (CICM 2012) in Bremen on July 11th 2012. The UITP workshop series aims at bringing together reasearchers interested in designing, developing and evaluating interfaces for interactive proof systems, such as theorem provers, formal method tools, and other tools manipulating and presenting mathematical formulae. Started in 1995, it can look back on seventeen years of history by now. The papers in the present volume give a good indication of the range of questions currently addressed in the UITP community; this ranges from interface design (Windsteiger; Dunchev et al) to using technologies such as machine learning to assist the user (Komendantskaya et al). The web features prominently (Tankink), and new technology necessitates changes right down to the very basic modes of interaction (Wenzel) - the old REPL (read, evaluate, print, loop) mode of interaction can not take advantage of modern technology, such as the web and multi-core machines.