Perception Learning: A Formal Separation of Sensory Representation Learning from Decision Learning

2025/10/28 by Sanyal, Suman
#Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.2510.24356

Abstract

We introduce Perception Learning (PeL), a paradigm that optimizes an agent's sensory interface f_ϕ:X\toZ using task-agnostic signals, decoupled from downstream decision learning g_θ:Z\toY. PeL directly targets label-free perceptual properties, such as stability to nuisances, informativeness without collapse, and controlled geometry, assessed via objective representation-invariant metrics. We formalize the separation of perception and decision, define perceptual properties independent of objectives or reparameterizations, and prove that PeL updates preserving sufficient invariants are orthogonal to Bayes task-risk gradients. Additionally, we provide a suite of task-agnostic evaluation metrics to certify perceptual quality.

Citations

• DINOv2: Learning Robust Visual Features without Supervision
• Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture
• data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language
• Masked Autoencoders Are Scalable Vision Learners
• Perceiver IO: A General Architecture for Structured Inputs & Outputs
• Perceiver: General Perception with Iterative Attention
• Learning Transferable Visual Models From Natural Language Supervision
• SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks
• Bootstrap your own latent: A new approach to self-supervised Learning
• A Simple Framework for Contrastive Learning of Visual Representations
• Momentum Contrast for Unsupervised Visual Representation Learning
• Momentum Contrast for Unsupervised Visual Representation Learning
• Representation Learning with Contrastive Predictive Coding
• World Models
• Representation Learning: A Review and New Perspectives
• Representation Learning: A Review and New Perspectives
• The information bottleneck method

Related

• Joint Perception and Control as Inference with an Object-based Implementation
• A Foundation to Perception Computing, Logic and Automata
• Artemis: Structured Visual Reasoning for Perception Policy Learning
• Perception-R1: Pioneering Perception Policy with Reinforcement Learning
• Using Human Perception to Regularize Transfer Learning