Multi-Task Learning

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Learningconcept0.827
  Inference of parameters encoding contingencies of the world (e.g., likelihood matrix A) at slower timescale than perception.
• Task balancing is still an open problem in multi-task learning.claim0.805
  Motivation for the proposed method.
• Learning performanceconcept0.788
  How well an RL agent learns, measured by reward curves and associated with causal emergence levels.
• Supervised Learningframework0.785
  Learning through physical changes in mechanical networks, as an example of learning outside neural systems.
• Meta-learningconcept0.785
  The capability of GPT-3 to learn tasks from few-shot prompts during runtime.
• Q-learningmethod0.782
  Model-free RL algorithm used in experimental comparison; employs ε-greedy exploration.
• Perceptual Learningconcept0.771
  Process of inferring causes of sensory information; unified with value learning as integral aspects of free energy minimization.
• Physical learningframework0.771
  Framework for solving inverse problems in which physical systems autonomously adapt their parameters in response to stimuli through local learning rules, without requiring computational design or explicit cost functions