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.

中文翻译：

---

开放世界持续学习：统一新奇检测和持续学习


随着 AI 代理越来越多地用于未知或新奇的真实开放世界，它们需要能够 （1） 识别 （a） 他们以前学过的物体，（b） 检测他们从未见过或学过的项目，以及 （2） 逐步学习新项目，以变得越来越博学和强大。（1） 称为新颖性检测或分布外 （OOD） 检测，（2） 称为类增量学习 （CIL），这是持续学习 （CL） 的一种设置。在现有的研究中，OOD 检测和 CIL 被认为是两个完全不同的问题。本文首先提供了一个理论证明，即对学习任务集中的每个任务进行良好的 OOD 检测（称为封闭世界 OOD 检测）是成功 CIL 的必要条件。我们通过将 CIL 分解为两个子问题来证明这一点：任务内预测 （WP） 和任务 ID 预测 （TP），并证明 TP 与封闭世界 OOD 检测相关。关键的理论结果是，无论 WP 和 OOD 检测（或 TP）是由 CIL 算法显式还是隐含地定义，良好的 WP 和良好的封闭世界 OOD 检测都是良好 CIL 的必要和充分条件，它统一了新颖性或 OOD 检测和持续学习（特别是 CIL）。我们将这种传统 CIL 称为封闭世界 CIL，因为它不会在开放世界中检测未来的 OOD 数据。然后，该论文证明该理论可以推广或扩展到开放世界 CIL，即提出的开放世界持续学习，它可以在开放世界中执行 CIL 并检测未来或开放世界的 OOD 数据。基于理论结果，还设计了新的 CIL 方法，这些方法在 CIL 准确性和连续 OOD 检测方面大大优于强大的基线。