Table of Links Abstract and 1 Introduction Abstract and 1 Introduction Related works


Problem setting


Methodology
4.1. Decision boundary-aware distillation
4.2. Knowledge consolidation


Experimental results and 5.1. Experiment Setup
5.2. Comparison with SOTA methods
5.3. Ablation study


Conclusion and future work and References Related works Related works Related works Problem setting Problem setting Problem setting Methodology
4.1. Decision boundary-aware distillation
4.2. Knowledge consolidation Methodology Methodology 4.1. Decision boundary-aware distillation 4.1. Decision boundary-aware distillation 4.2. Knowledge consolidation 4.2. Knowledge consolidation Experimental results and 5.1. Experiment Setup
5.2. Comparison with SOTA methods
5.3. Ablation study Experimental results and 5.1. Experiment Setup Experimental results and 5.1. Experiment Setup 5.2. Comparison with SOTA methods 5.2. Comparison with SOTA methods 5.3. Ablation study 5.3. Ablation study Conclusion and future work and References Conclusion and future work and References Conclusion and future work and References Supplementary Material Supplementary Material Details of the theoretical analysis on KCEMA mechanism in IIL
Algorithm overview
Dataset details
Implementation details
Visualization of dusted input images
More experimental results Details of the theoretical analysis on KCEMA mechanism in IIL Details of the theoretical analysis on KCEMA mechanism in IIL Algorithm overview Algorithm overview Dataset details Dataset details Implementation details Implementation details Visualization of dusted input images Visualization of dusted input images More experimental results More experimental results 5.2. Comparison with SOTA methods Tab. 1 shows the test performance of different methods on the Cifar-100 and ImageNet-100. The proposed method achieves the best performance promotion after ten consecutive IIL tasks by a large margin with a low forgetting rate. Although ISL [13] which is proposed for a similar setting of learning from new sub-categories has a low forgetting rate, it fails on the new requirement of model enhancement. Attain a better performance on the test data is more important than forgetting on a certain data. In the new IIL setting, all rehearsal-based methods including iCarl [22], PODNet [4], Der [31] and OnPro [29], not perform well. Old exemplars can cause memory overfitting and model bias [35]. Thus, limited old exemplars not always have a positive influence to the stability and plasticity [26], especially in the IIL task. Forgetting rate of rehearsal-based methods is high compared to other methods, which also explains their performance degradation on the test data. Detailed performance at each learning phase is shown in Fig. 4. Compared to other methods that struggle in resisting forgetting, our method is the only one that stably promotes the existing model on both of the two datasets. Following ISL [13], we further apply our method on the incremental sub-population learning as shown in Tab. 2. Sub-population incremental learning is a special case of the IIL where new knowledge comes from the new subclasses. Compared to the SOTA ISL [13], our method is notably superior in learning new subclasses over long incremental steps with a comparable small forgetting rate. Noteworthy, ISL [13] use Continual Hyperparameter Framework (CHF) [3] searching the best learning rate (such as low to 0.005 in 15-step task) for each setting. While our method learns utilizing ISL pretrained base model with a fixed learning rate (0.05). Low learning rate in ISL reduces the forgetting but hinders the new knowledge learning. The proposed method well balances learning new from unseen subclasses and resisting forgetting on seen classes. Authors:
(1) Qiang Nie, Hong Kong University of Science and Technology (Guangzhou);
(2) Weifu Fu, Tencent Youtu Lab;
(3) Yuhuan Lin, Tencent Youtu Lab;
(4) Jialin Li, Tencent Youtu Lab;
(5) Yifeng Zhou, Tencent Youtu Lab;
(6) Yong Liu, Tencent Youtu Lab;
(7) Qiang Nie, Hong Kong University of Science and Technology (Guangzhou);
(8) Chengjie Wang, Tencent Youtu Lab. Authors: Authors: (1) Qiang Nie, Hong Kong University of Science and Technology (Guangzhou); (2) Weifu Fu, Tencent Youtu Lab; (3) Yuhuan Lin, Tencent Youtu Lab; (4) Jialin Li, Tencent Youtu Lab; (5) Yifeng Zhou, Tencent Youtu Lab; (6) Yong Liu, Tencent Youtu Lab; (7) Qiang Nie, Hong Kong University of Science and Technology (Guangzhou); (8) Chengjie Wang, Tencent Youtu Lab. This paper is available on arxiv under CC BY-NC-ND 4.0 Deed (Attribution-Noncommercial-Noderivs 4.0 International) license. This paper is available on arxiv under CC BY-NC-ND 4.0 Deed (Attribution-Noncommercial-Noderivs 4.0 International) license. available on arxiv available on arxiv

Part of HackerNoon's growing list of open-source research papers, promoting free access to academic material.

SAGE Net Ablation Study: Analyzing the Impact of Input Sequence Length on Performance

About Author

Comments

TOPICS

THIS ARTICLE WAS FEATURED IN

Related Stories

Untitled Story

A Coarse-to-Fine Approach to Text-Guided 3D Localization Without Ground Truth

Understanding GAN Mode Collapse: Causes and Solutions

Data Scarcity Solution: S-CycleGAN for CT-to-Ultrasound Translation

Incremental Learning: Comparing Methods for Catastrophic Forgetting and Model Promotion

Optimizing SAGE Net: Achieving High Performance with Shorter Input Sequences for Online Inference

A Coarse-to-Fine Approach to Text-Guided 3D Localization Without Ground Truth

Understanding GAN Mode Collapse: Causes and Solutions

Data Scarcity Solution: S-CycleGAN for CT-to-Ultrasound Translation

Incremental Learning: Comparing Methods for Catastrophic Forgetting and Model Promotion

Optimizing SAGE Net: Achieving High Performance with Shorter Input Sequences for Online Inference

Light-Mode

Classic

Newspaper

Dark-Mode

Neon Noir

Minty

HN StartUps