Search Results for author:
Found 9 papers, 4 papers with code
Multi-agent cooperation through learning-aware policy gradients
Self-interested individuals often fail to cooperate, posing a fundamental challenge for multi-agent learning.
A Complexity-Based Theory of Compositionality
Here, we propose such a definition, which we call representational compositionality, that accounts for and extends our intuitions about compositionality.
In-context learning and Occam's razor
While the No Free Lunch Theorem states that we cannot obtain theoretical guarantees for generalization without further assumptions, in practice we observe that simple models which explain the training data generalize best: a principle called Occam's razor.
Does learning the right latent variables necessarily improve in-context learning?
Our study highlights the intrinsic limitations of Transformers in achieving structured ICL solutions that generalize, and shows that while inferring the right latents aids interpretability, it is not sufficient to alleviate this problem.
Amortizing intractable inference in large language models
Autoregressive large language models (LLMs) compress knowledge from their training data through next-token conditional distributions.
Discrete, compositional, and symbolic representations through attractor dynamics
Symbolic systems are powerful frameworks for modeling cognitive processes as they encapsulate the rules and relationships fundamental to many aspects of human reasoning and behavior.
Consciousness in Artificial Intelligence: Insights from the Science of Consciousness
From these theories we derive "indicator properties" of consciousness, elucidated in computational terms that allow us to assess AI systems for these properties.
Sources of Richness and Ineffability for Phenomenally Conscious States
Conscious states (states that there is something it is like to be in) seem both rich or full of detail, and ineffable or hard to fully describe or recall.
Visual representations derived from multiplicative interactions
Biological sensory systems appear to rely on canonical nonlinear computations that can be readily adapted to a broad range of representational objectives.
