Toward industrial use of continual learning : new metrics proposal for class incremental learning
This work addresses the need for better evaluation metrics in continual learning to support industrial adoption, though it is incremental as it focuses on improving existing assessment frameworks rather than introducing new learning methods.
The paper tackles the problem of evaluating continual learning methods in class incremental learning by showing that existing metrics like mean task accuracy are misleading and overly optimistic for industrial applications. It proposes two new metrics, Minimal Incremental Class Accuracy (MICA) and a derived scalar metric, to provide fairer and more useful comparisons of method performance.
In this paper, we investigate continual learning performance metrics used in class incremental learning strategies for continual learning (CL) using some high performing methods. We investigate especially mean task accuracy. First, we show that it lacks of expressiveness through some simple experiments to capture performance. We show that monitoring average tasks performance is over optimistic and can lead to misleading conclusions for future real life industrial uses. Then, we propose first a simple metric, Minimal Incremental Class Accuracy (MICA) which gives a fair and more useful evaluation of different continual learning methods. Moreover, in order to provide a simple way to easily compare different methods performance in continual learning, we derive another single scalar metric that take into account the learning performance variation as well as our newly introduced metric.