XLDA: Linear Discriminant Analysis for Scaling Continual Learning to Extreme Classification at the Edge
This enables efficient continual learning on resource-constrained edge devices for applications like large-scale image recognition.
The paper tackles the problem of scaling class-incremental learning to extreme classification scenarios with up to 81k classes for edge deployment, showing up to 42x training speedup and 5x inference speedup.
Streaming Linear Discriminant Analysis (LDA) while proven in Class-incremental Learning deployments at the edge with limited classes (upto 1000), has not been proven for deployment in extreme classification scenarios. In this paper, we present: (a) XLDA, a framework for Class-IL in edge deployment where LDA classifier is proven to be equivalent to FC layer including in extreme classification scenarios, and (b) optimizations to enable XLDA-based training and inference for edge deployment where there is a constraint on available compute resources. We show up to 42x speed up using a batched training approach and up to 5x inference speedup with nearest neighbor search on extreme datasets like AliProducts (50k classes) and Google Landmarks V2 (81k classes)