Benchmarks for Physical Reasoning AI
A survey of physical reasoning benchmarks and approaches in artificial intelligence.
- RUB
- Computer Science
- INI
- People
- Dr. Robin Schiewer
-
Exploring the limits of hierarchical world models in reinforcement learningSchiewer, R., Subramoney, A., & Wiskott, L.Scientific Reports, 14(1)Hebbian Descent: A Unified View on Log-Likelihood LearningMelchior, J., Schiewer, R., & Wiskott, L.Neural Computation, 36(9), 1669–1712
2023
Benchmarks for Physical Reasoning AIMelnik, A., Schiewer, R., Lange, M., Muresanu, A. I., Saeidi, M., Garg, A., & Ritter, H.Transactions on Machine Learning Research2022
Latent Representation Prediction NetworksHlynsson, H. D., Schüler, M., Schiewer, R., Glasmachers, T., & Wiskott, L.International Journal of Pattern Recognition and Artificial Intelligence, 36(01), 2251002Modular Networks Prevent Catastrophic Interference in Model-Based Multi-task Reinforcement LearningSchiewer, R., & Wiskott, L.In Machine Learning, Optimization, and Data Science (pp. 299–313) Springer International Publishing2020
Latent Representation Prediction NetworksWinter Term 2022/2023
Project seminar Machine Learning: Unsupervised Methods (with Problem Based Learning) Summer Term 2022
Lab courses Introduction to Python Winter Term 2021/2022
Lectures Machine Learning: Unsupervised Methods (with Problem Based Learning) Summer Term 2021
Lab courses Introduction to Python Winter Term 2020/2021
Lectures Machine Learning: Unsupervised Methods (with Problem Based Learning) Seminars Master Seminar: Methods of Modern Reinforcement Learning Summer Term 2020
Lab courses Introduction to Python Seminars Master Seminar: Methods of Modern Reinforcement Learning Summer Term 2019
Lab courses Introduction to Python
2022 
2019 Predicting Latent Space Representations for Planning
Pairing the EfferenceNet with a good but generic feature map allows us to perform an accurate search in the latent space of manipulating unseen objects. This remarkably simple method, inspired by the neurology of the cerebellum, reveals a promising line of future work. We validate our method by on a viewpoint-matching task derived from the NORB data set.
2019 Combat Task Interference in Multi-Task Model-Based Reinforcement Learning by Using Separate Dynamics Models
In model-free multi-task reinforcement learning (RL), abundant work shows that a shared policy network can improve performance across the different tasks. The rationale behind this is that an agent can learn similarities that all tasks have in common and thus effectively enrich the sample count for all tasks at hand. In model-based multi-task RL however, we found evidence suggesting that a dynamics model can suffer from task confusion or catastrophic interference if it is trained on multiple tasks at once.
The Institut für Neuroinformatik (INI) is a research unit of the Faculty of Computer Science at the Ruhr-Universität Bochum. Its scientific goal is to understand the fundamental principles through which organisms generate behavior and cognition while linked to their environments through sensory and effector systems. Inspired by our insights into such natural cognitive systems, we seek new solutions to problems of information processing in artificial cognitive systems. We draw from a variety of disciplines that include experimental psychology and neurophysiology as well as machine learning, neural artificial intelligence, computer vision, and robotics.
Universitätsstr. 150, Building NB, Room 3/32
D-44801 Bochum, GermanyTel: (+49) 234 32-28967
Fax: (+49) 234 32-14210