Research map / Continual learning / Online and replay-based
GDumb: A Simple Approach that Questions Our Progress in Continual Learning
Ameya Prabhu, Philip H. S. Torr, Puneet K. Dokania
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use GDumb instead of ER-MIR | continual learner · Split-CIFAR10 A1 (k=1000) | accuracy: 47.6 -> 61.3 | Improved |
| use GDumb instead of ER-MIR | continual learner · Split-MNIST A1 (k=500) | accuracy: 87.6 -> 91.9 | Improved |
| use GDumb instead of GMED (MIR) | continual learner · Split-CIFAR10 A2 (k=500) | accuracy: 35.5 -> 45.8 | Improved |
| use GDumb instead of ARM | continual learner · Split-CIFAR10 A3 | accuracy: 26.4 -> 61.3 | Improved |
| increase memory from 200 to 1K (ablation) | GDumb · Split-CIFAR10 A1 | accuracy: 35.0 -> 61.3 | Improved |
| increase memory from 200 to 1K | ER-MIR · Split-CIFAR10 A1 | accuracy: 29.8 -> 47.6 | Improved |
| use GDumb instead of iCARL/BiC with 20 tasks | continual learner · CIFAR100 B2 | accuracy: 63.8 -> 45.2 | Hurt |
| use GDumb instead of UCIR (CNN) with 50 tasks | continual learner · CIFAR100 B3 | accuracy: 49.3 -> 58.4 | Improved |
| use GDumb (CNN) instead of PODNet (CNN) | continual learner · CIFAR100 B3 | accuracy: 58.0 -> 58.4 | No clear effect |
| use GDumb instead of iTAML | continual learner · MNIST B1 | accuracy: 97.9 -> 97.8 | No clear effect |
| use GDumb instead of iCARL with k=4500 | continual learner · Split-TinyImageNet C2 | accuracy: 48.55 -> 45.50 | Hurt |
| use GDumb instead of iCARL with k=9000 | continual learner · Split-TinyImageNet C2 | accuracy: 49.94 -> 57.27 | Improved |
Later papers that compared with this method
13 later accepted papers report GDumb as a baseline in their results tables; 10 of them report a better number on most of the results they share with it.
- CRAM: Large Scale Video Continual Learning with Bootstrapped Compression (ICCV 2025 · better on 3 of 3 shared results)
- Orchestrate Latent Expertise: Advancing Online Continual Learning with Multi-Level Supervision and Reverse Self-Distillation (CVPR 2024 · better on 0 of 3 shared results)
- RanDumb: Random Representations Outperform Online Continually Learned Representations (NeurIPS 2024 · better on 0 of 1 shared result)
- First Session Adaptation: A Strong Replay-Free Baseline for Class-Incremental Learning (ICCV 2023 · better on 2 of 3 shared results)
- Label-Efficient Online Continual Object Detection in Streaming Video (ICCV 2023 · better on 2 of 2 shared results)
- Online Noisy Continual Relation Learning (AAAI 2023 · better on 1 of 1 shared result)
- Batch Model Consolidation: A Multi-Task Model Consolidation Framework (CVPR 2023 · better on 1 of 1 shared result)
- Dealing With Cross-Task Class Discrimination in Online Continual Learning (CVPR 2023 · better on 1 of 1 shared result)
5 more not listed.
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 2 on most shared results.
- iCARL — iCaRL: Incremental Classifier and Representation Learning (CVPR 2017) · better on 1 of 3 shared results
- iTAML — iTAML: An Incremental Task-Agnostic Meta-learning Approach (CVPR 2020) · better on 0 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- A continual learning survey: Defying forgetting in classification tasks (TPAMI 2021 · similarity 0.90)
- LoRanPAC: Low-rank Random Features and Pre-trained Models for Bridging Theory and Practice in Continual Learning (ICLR 2025 · similarity 0.89)
- GACL: Exemplar-Free Generalized Analytic Continual Learning (NeurIPS 2024 · similarity 0.87)
- Convergence and Implicit Bias of Gradient Descent on Continual Linear Classification (ICLR 2025 · similarity 0.87)
- CLDyB: Towards Dynamic Benchmarking for Continual Learning with Pre-trained Models (ICLR 2025 · similarity 0.87)
- Forgetting, Ignorance or Myopia: Revisiting Key Challenges in Online Continual Learning (NeurIPS 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
