This paper presents OpenRoIS, a community-driven open-source middleware providing a concrete implementation of the RoIS Framework 2.0, taking the position that an openly developed, paradigm-neutral implementation is what carries the standard from specification to practice.
Sebastian Carrera Villalobos, Christopher Nolan Arellano, Arne Hitzmann et al.· 0 citations
This work formalizes RL safety with a safety-critical Markov decision process (SC-MDP), which requires categorical, rather than high-probability, avoidance of catastrophic failure states and provides a simple, principled recipe for safe RL: train and deploy RL policies with the most permissive safety filter that is ava...
Donggeon David Oh, D. Nguyen, Hai-Min Hu et al.· Proceedings of IASEAI Confer...· 5 citations
This work introduces a novel inductive bias towards simple policies in reinforcement learning by minimizing the entropy of entire action trajectories, corresponding to the number of bits required to describe information in action trajectories after the agent observes state trajectories.
Bang You, Chenxu Wang, Wen-Ju Yang et al.· 0 citations
This paper investigates how to learn representations by maximizing the rollout total correlation, the correlation among all learned representations and actions within the trajectories produced by the agent, and proposes an intrinsic reward based on the learned representation for better exploration.
Bang You, Hua-Ping Liu, Jan Peters et al.· Trans. Mach. Learn. Res.· 3 citations· ⚡1
Reach audiences
Advertise in front of researchers, engineers, and readers.
PARTS (Policy Adaptation with RL on Targeted Subtasks), a real-world subtask RL framework that concentrates practice at bottlenecks while allowing training rollouts to proceed with minimal human intervention, is presented.
Sichang Su, B. Yang, Zhi-Yun Deng et al.· 0 citations
The proposed epistemic uncertainty-driven adaptive rollout strategy for offline world model training following an auto-curriculum training scheme indicates that epistemic uncertainty is useful not only for downstream policy regularization, but also for making world model training itself more compute-efficient.
Nikodem Sebastian Zymla, L. Thiele, Johannes Pitz· 0 citations
FootQuery is presented, a perceptive locomotion framework that queries depth history using each foot's predicted next touchdown using each foot's predicted next touchdown to organize visual history around anticipated contacts for perceptive humanoid locomotion.
A Diffusion-based Uncertainty-aware Optimization (DUO) algorithm that learns demonstration-free crawling controllers for a simulated, muscle-actuated CyberOctopus, which represents the first application of diffusion-based control to soft multi-arm robots in contact-rich simulations.
Seung Hyun Kim, Heng-Sheng Chang, Kimia Kazemi et al.· 1 citation
This work designs each task curriculum with compositional tasks that combine aspects of the tasks seen in the sequence, and factorise this composition along the axes of action and perception to better understand how different input modalities bottleneck knowledge reuse.
Hao-Yu Zhou, Joe Watson, Anson Lei et al.· 0 citations
A systematic comparison of two state-of-the-art motion-imitation reinforcement learning (MIRL) pipelines, one built on SCONE/HyFyDy and one built on MuJoCo/MyoSim, concludes that the more advanced physiological realism of HyFyDy currently makes it more suitable for musculoskeletal modeling.
Ayah Ahmad, Claire E. Borden, Maegan Tucker· 0 citations
This work proposes an alternative data processing method for remote sensing applications via the use of data in Comma-Seperated Value format that shows a reduction of processing time, an improvement in accurate object detection by Machine Learning, and an increase in SNR (Signal-to-noise ratio), PSNR (Peak signal-to-no...
Recent advances in large language models have improved their effectiveness as back-end components for voice assistants, particularly in intent understanding and context-aware input classification. However, online-hosted models introduce network dependency and variable inference latency, limiting their suitability for t...
Daniel Henel, Frederik Werner, Alexander Langmann et al.· 0 citations
A new system identifies and corrects unreliable object descriptions stored in a robot’s memory, improving the robot’s ability to interact with its environment.
Robots are getting smarter, but how can their hardware match that growth? New Microsoft Research findings show that moving AI inference beyond the robot can improve task success, boost efficiency, and support more advanced physical AI workloads. The post Offloaded inference for real-world physical AI robotics appeared first on Microsoft Research.
