2023/03/30 by Tyler Malloy, Miao Liu, Malloy, Tailia +9 · 3 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · Mathematics · Neuroscience · Psychology · #Artificial Intelligence (cs.AI) #Artificial intelligence #Autoencoder #Cell Image Analysis Techniques #Cognitive psychology #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Deep learning #Encoder #FOS: Biological sciences #FOS: Computer and information sciences #Feature learning #Generalization #Human-Computer Interaction (cs.HC) #Machine learning #Mathematics #Neural Networks and Applications #Neural dynamics and brain function #Neurons and Cognition (q-bio.NC) #Perception #Psychology #Reinforcement learning #Representation (politics) #Robustness (evolution) #Task (project management) #Visual learning
paper · pdf · doi:10.48550/arxiv.2303.17508
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2023/03/30 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Humans learn quickly even in tasks that contain complex visual information. This is due in part to the efficient formation of compressed representations of visual information, allowing for better generalization and robustness. However, compressed representations alone are insufficient for explaining the high speed of human learning. Reinforcement learning (RL) models that seek to replicate this impressive efficiency may do so through the use of factored representations of tasks. These informationally simplistic representations of tasks are similarly motivated as the use of compressed representations of visual information. Recent studies have connected biological visual perception to disentangled and compressed representations. This raises the question of how humans learn to efficiently represent visual information in a manner useful for learning tasks. In this paper we present a model of human factored representation learning based on an altered form of a β-Variational Auto-encoder used in a visual learning task. Modelling results demonstrate a trade-off in the informational complexity of model latent dimension spaces, between the speed of learning and the accuracy of reconstructions.