Search Results for author:
Found 8 papers, 3 papers with code
Rapid Open-World Adaptation by Adaptation Principles Learning
This is an important characteristic of intelligent agents, as it allows them to continue to function effectively in novel or unexpected situations, but still stands as a critical challenge for deep reinforcement learning (DRL).
Efficient Open-world Reinforcement Learning via Knowledge Distillation and Autonomous Rule Discovery
In this work, we propose a general framework that is applicable to deep reinforcement learning agents.
NovPhy: A Testbed for Physical Reasoning in Open-world Environments
But is it enough to only have physical reasoning capabilities to operate in a real physical environment?
Don't do it: Safer Reinforcement Learning With Rule-based Guidance
During training, reinforcement learning systems interact with the world without considering the safety of their actions.
Measuring Difficulty of Novelty Reaction
In this paper, we propose to define the novelty reaction difficulty as a relative difficulty of performing the known task after the introduction of the novelty.
Phy-Q as a measure for physical reasoning intelligence
Inspired by how human IQ is calculated, we define the physical reasoning quotient (Phy-Q score) that reflects the physical reasoning intelligence of an agent using the physical scenarios we considered.
Deep Q-Network for Angry Birds
Angry Birds is a popular video game in which the player is provided with a sequence of birds to shoot from a slingshot.
Learning Underlying Physical Properties From Observations For Trajectory Prediction
We show that by using physical laws together with deep learning we achieve a better human-interpretability of learned physical properties, transfer of knowledge to a game with similar physics and very accurate predictions for previously unseen data.
