Non-Markov Decision Processes and Reinforcement Learning
We present non-Markov decision processes, where rewards and dynamics can depend on the history of events. This is contrast with Markov Decision Processes, where the dependency is limited to the last state and action.
Augmentation-free unsupervised approach for point clouds
Unsupervised learning on 3D point clouds has undergone a rapid evolution, especially thanks to data augmentation-based contrastive methods.
Algorithmic Impact Assessment tool
The Algorithmic Impact Assessment (AIA) is a mandatory risk assessment tool intended to support the Treasury Board’s Directive on Automated Decision-Making.
ALIGNER Fundamental Rights Impact Assessment
Artificial intelligence can enhance law enforcement agencies’ capabilities to prevent, investigate, detect and prosecute crimes, as well as to predict and anticipate them.
A General Language Assistant as a Laboratory for Alignment
Given the broad capabilities of large language models, it should be possible to work towards a general-purpose, text-based assistant that is aligned with human values, meaning that it is helpful, honest, and harmless.