Slumbering to Precision: Enhancing Artificial Neural Network Calibration Through Sleep-like Processes
This work provides a novel method for improving the trustworthiness of confidence estimates in deep neural networks, which is a critical problem for AI systems deployed in real-world applications.
This paper addresses the issue of overconfident artificial neural networks by introducing Sleep Replay Consolidation (SRC), a post-training, sleep-like phase that selectively replays internal representations to update network weights. SRC improves calibration without supervised retraining and, when combined with temperature scaling, achieves the best Brier score and entropy trade-offs for AlexNet and VGG19.
Artificial neural networks are often overconfident, undermining trust because their predicted probabilities do not match actual accuracy. Inspired by biological sleep and the role of spontaneous replay in memory and learning, we introduce Sleep Replay Consolidation (SRC), a novel calibration approach. SRC is a post-training, sleep-like phase that selectively replays internal representations to update network weights and improve calibration without supervised retraining. Across multiple experiments, SRC is competitive with and complementary to standard approaches such as temperature scaling. Combining SRC with temperature scaling achieves the best Brier score and entropy trade-offs for AlexNet and VGG19. These results show that SRC provides a fundamentally novel approach to improving neural network calibration. SRC-based calibration offers a practical path toward more trustworthy confidence estimates and narrows the gap between human-like uncertainty handling and modern deep networks.