Abstract:As AI agents are increasingly used in the real open world with unknowns or novelties, they need the ability to (1) recognize objects that (a) they have learned before and (b) detect items that they have never seen or learned, and (2) learn the new items incrementally to become more and more knowledgeable and powerful. (1) is called novelty detection or out-of-distribution (OOD) detection and (2) is called class incremental learning (CIL), which is a setting of continual learning (CL). In existing research, OOD detection and CIL are regarded as two completely different problems. This paper first provides a theoretical proof that good OOD detection for each task within the set of learned tasks (called closed-world OOD detection) is necessary for successful CIL. We show this by decomposing CIL into two sub-problems: within-task prediction (WP) and task-id prediction (TP), and proving that TP is correlated with closed-world OOD detection. The key theoretical result is that regardless of whether WP and OOD detection (or TP) are defined explicitly or implicitly by a CIL algorithm, good WP and good closed-world OOD detection are necessary and sufficient conditions for good CIL, which unifies novelty or OOD detection and continual learning (CIL, in particular). We call this traditional CIL the closed-world CIL as it does not detect future OOD data in the open world. The paper then proves that the theory can be generalized or extended to open-world CIL, which is the proposed open-world continual learning, that can perform CIL in the open world and detect future or open-world OOD data. Based on the theoretical results, new CIL methods are also designed, which outperform strong baselines in CIL accuracy and in continual OOD detection by a large margin.

Self-Supervised Features Improve Open-World Learning

Open World Classification with Adaptive Negative Samples.

A Review of Open-World Learning and Steps Toward Open-World Learning Without Labels

Robust Semi-Supervised Learning for Self-learning Open-World Classes

PSS: Progressive Sample Selection for Open-World Visual Representation Learning

A Graph-Theoretic Framework for Understanding Open-World Semi-Supervised Learning

Self-Initiated Open World Learning for Autonomous AI Agents

Rethinking Open-World Semi-Supervised Learning: Distribution Mismatch and Inductive Inference

Deep Active Learning in the Open World

Open-world Machine Learning: A Review and New Outlooks

Dynamic Against Dynamic: An Open-set Self-learning Framework

From Open Vocabulary to Open World: Teaching Vision Language Models to Detect Novel Objects

Detecting and Learning Out-of-Distribution Data in the Open world: Algorithm and Theory

Learning refined features for open-world text classification with class description and commonsense knowledge

Unsupervised Recognition of Unknown Objects for Open-World Object Detection

Open-World Continual Learning: Unifying Novelty Detection and Continual Learning

Open World Semi-supervised Learning Based on Multi-scale Enhanced Feature

Towards Open World Recognition

Rethinking Open-World Object Detection in Autonomous Driving Scenarios

Label Distribution-based Open-world Semi-supervised Learning

Targeted Representation Alignment for Open-World Semi-Supervised Learning