Publication
AI-KD: Adversarial learning and Implicit regularization for self-Knowledge Distillation
Hyungmin Kim; Sungho Suh; Sunghyun Baek; Daehwan Kim; Daun Jeong; Hansang Cho; Junmo Kim
In: Knowledge-Based Systems (KBS), Vol. 293, Page 111692, Elsevier, 2024.
Abstract
We present a novel adversarial penalized self-knowledge distillation method, named adversarial learning and implicit regularization for self-knowledge distillation (AI-KD), which regularizes the training procedure by adversarial learning and implicit distillations. Our model not only distills the deterministic and progressive knowledge which are from the pre-trained and previous epoch predictive probabilities but also transfers the knowledge of the deterministic predictive distributions using adversarial learning. The motivation is that the self-knowledge distillation methods regularize the predictive probabilities with soft targets, but the exact distributions may be hard to predict. Our proposed method deploys a discriminator to distinguish the distributions between the pre-trained and student models while the student model is trained to fool the discriminator in the trained procedure. Thus, the student model not only can learn the pre-trained model’s predictive probabilities but also align the distributions between the pre-trained and student models. We demonstrate the effectiveness of the proposed method with network architectures on multiple datasets and show the proposed method achieves better performance than existing approaches.