CONCLAD: COntinuous Novel CLAss Detector
This addresses the unrealistic reliance on oracles in continual learning, offering a practical solution for incremental novelty detection.
The paper tackles the problem of continual novel class detection in post-deployment data, introducing CONCLAD, which uses uncertainty estimation and pseudo-labeling to automatically detect and update for novel classes, achieving effective separation across multiple datasets.
In the field of continual learning, relying on so-called oracles for novelty detection is commonplace albeit unrealistic. This paper introduces CONCLAD ("COntinuous Novel CLAss Detector"), a comprehensive solution to the under-explored problem of continual novel class detection in post-deployment data. At each new task, our approach employs an iterative uncertainty estimation algorithm to differentiate between known and novel class(es) samples, and to further discriminate between the different novel classes themselves. Samples predicted to be from a novel class with high-confidence are automatically pseudo-labeled and used to update our model. Simultaneously, a tiny supervision budget is used to iteratively query ambiguous novel class predictions, which are also used during update. Evaluation across multiple datasets, ablations and experimental settings demonstrate our method's effectiveness at separating novel and old class samples continuously. We will release our code upon acceptance.