Benchmarks for Physical Reasoning AI
A survey of physical reasoning benchmarks and approaches in artificial intelligence.
@article{SchiewerSubramoneyWiskott2024, author = {Schiewer, Robin and Subramoney, Anand and Wiskott, Laurenz}, title = {Exploring the limits of hierarchical world models in reinforcement learning}, journal = {Scientific Reports}, volume = {14}, number = {1}, month = {November}, year = {2024}, doi = {10.1038/s41598-024-76719-w}, }
@article{MelchiorSchiewerWiskott2024, author = {Melchior, Jan and Schiewer, Robin and Wiskott, Laurenz}, title = {Hebbian Descent: A Unified View on Log-Likelihood Learning}, journal = {Neural Computation}, volume = {36}, number = {9}, pages = {1669–1712}, month = {August}, year = {2024}, doi = {10.1162/neco_a_01684}, }
@article{MelnikSchiewerLangeEtAl2023, author = {Melnik, Andrew and Schiewer, Robin and Lange, Moritz and Muresanu, Andrei Ioan and Saeidi, Mozhgan and Garg, Animesh and Ritter, Helge}, title = {Benchmarks for Physical Reasoning AI}, journal = {Transactions on Machine Learning Research}, year = {2023}, }
@article{HlynssonSchülerSchiewerEtAl2022, author = {Hlynsson, Hlynur David and Schüler, Merlin and Schiewer, Robin and Glasmachers, Tobias and Wiskott, Laurenz}, title = {Latent Representation Prediction Networks}, journal = {International Journal of Pattern Recognition and Artificial Intelligence}, volume = {36}, number = {01}, pages = {2251002}, year = {2022}, doi = {10.1142/S0218001422510028}, }
@incollection{SchiewerWiskott2022, author = {Schiewer, Robin and Wiskott, Laurenz}, title = {Modular Networks Prevent Catastrophic Interference in Model-Based Multi-task Reinforcement Learning}, booktitle = {Machine Learning, Optimization, and Data Science}, pages = {299–313}, publisher = {Springer International Publishing}, year = {2022}, doi = {10.1007/978-3-030-95470-3_23}, }
@article{HlynssonSchülerSchiewerEtAl2020, author = {Hlynsson, Hlynur Davíð and Schüler, Merlin and Schiewer, Robin and Glasmachers, Tobias and Wiskott, Laurenz}, title = {Latent Representation Prediction Networks}, journal = {arXiv preprint arXiv:2009.09439}, year = {2020}, }
Project seminar | Machine Learning: Unsupervised Methods (with Problem Based Learning) |
Lab courses | Introduction to Python |
Lectures | Machine Learning: Unsupervised Methods (with Problem Based Learning) |
Lab courses | Introduction to Python |
Lectures | Machine Learning: Unsupervised Methods (with Problem Based Learning) |
Seminars | Master Seminar: Methods of Modern Reinforcement Learning |
Lab courses | Introduction to Python |
Seminars | Master Seminar: Methods of Modern Reinforcement Learning |
Lab courses | Introduction to Python |
A survey of physical reasoning benchmarks and approaches in artificial intelligence.
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.
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, Germany
Tel: (+49) 234 32-28967
Fax: (+49) 234 32-14210