Online Continual Learning via Multiple Deep Metric Learning and Uncertainty-guided Episodic Memory Replay -- 3rd Place Solution for ICCV 2021 Workshop SSLAD Track 3A Continual Object Classification

2021-11-04 · Muhammad Rifki Kurniawan, Xing Wei, Yihong Gong

Online continual learning in the wild is a very difficult task in machine learning. Non-stationarity in online continual learning potentially brings about catastrophic forgetting in neural networks. Specifically, online continual learning for autonomous driving with SODA10M dataset exhibits extra problems on extremely long-tailed distribution with continuous distribution shift. To address these problems, we propose multiple deep metric representation learning via both contrastive and supervised contrastive learning alongside soft labels distillation to improve model generalization. Moreover, we exploit modified class-balanced focal loss for sensitive penalization in class imbalanced and hard-easy samples. We also store some samples under guidance of uncertainty metric for rehearsal and perform online and periodical memory updates. Our proposed method achieves considerable generalization with average mean class accuracy (AMCA) 64.01% on validation and 64.53% AMCA on test set.

📄 PDF Abstract BibTeX arXiv:2111.02757

Code (1)

mrifkikurniawan/sslad 공식 구현 pytorch

Tasks

Autonomous Driving Continual Learning Contrastive Learning Metric Learning Representation Learning

Methods 이 논문이 사용한 방법론

Contrastive Learning 설명 없음

Focal Loss A Focal Loss function addresses class imbalance during training in tasks like object detection. Focal loss applies a modulating term to the cross entropy loss in order to…

Uncertainty Quantification in Continual Open-World Learning

2024-12-21 · Amanda S. Rios, Ibrahima J. Ndiour, Parual Datta, Jaroslaw Sydir 외

AI deployed in the real-world should be capable of autonomously adapting to novelties encountered after deployment. Yet, in the field of continual learning, the reliance on novelty and labeling oracles is commonplace alb…

Active Learning AI Agent Continual Learning Novelty Detection+1

Active Neural Mapping

2023-08-30 · ICCV 2023 1 · Zike Yan, Haoxiang Yang, Hongbin Zha

We address the problem of active mapping with a continually-learned neural scene representation, namely Active Neural Mapping. The key lies in actively finding the target space to be explored with efficient agent movemen…

Continual uncertainty learning

2026-02-19 · Heisei Yonezawa, Ansei Yonezawa, Itsuro Kajiwara arxiv

Robust control of mechanical systems with multiple uncertainties remains a fundamental challenge, particularly when nonlinear dynamics and operating-condition variations are intricately intertwined. Although deep reinfor…

Reinforcement Learning Continual Learning

Uncertainty-guided Continual Learning with Bayesian Neural Networks

2019-06-06 · ICLR 2020 1 · Sayna Ebrahimi, Mohamed Elhoseiny, Trevor Darrell, Marcus Rohrbach

Continual learning aims to learn new tasks without forgetting previously learned ones. This is especially challenging when one cannot access data from previous tasks and when the model has a fixed capacity. Current regul…

Continual Learning

RECALL: Recovery Experience Collection for Active Lifelong Learning in Vision-Language-Action Models

2026-06-22 · Ulas Berk Karli, Tesca Fitzgerald arxiv

Vision-Language-Action (VLA) models are commonly fine-tuned through passive imitation learning, where additional demonstrations are collected for tasks where the policy performs poorly. This approach incurs several downs…