We introduce Perception Learning (PeL), a paradigm that optimizes an agent's sensory interface $f_\phi:\mathcal{X}\to\mathcal{Z}$ using task-agnostic signals, decoupled from downstream decision learning $g_\theta:\mathcal{Z}\to\mathcal{Y}$. 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.

翻译：本文提出感知学习（Perception Learning，PeL）这一范式，它通过任务无关信号优化智能体的感官接口 $f_\phi:\mathcal{X}\to\mathcal{Z}$，并将其与下游决策学习 $g_\theta:\mathcal{Z}\to\mathcal{Y}$ 解耦。PeL 直接针对无标签的感知特性进行优化，例如对干扰因素的稳定性、无坍缩的信息丰富性以及可控的几何结构，这些特性通过客观的表征不变性度量进行评估。我们形式化地分离了感知与决策过程，定义了独立于具体目标或重参数化的感知特性，并证明保持充分不变性的 PeL 更新与贝叶斯任务风险梯度正交。此外，我们提供了一套任务无关的评估指标来验证感知质量。