# Lifelong Learning Without a Task Oracle

arXiv: https://arxiv.org/abs/2011.04783 · HTML: https://mlsanity.com/paper/2011.04783/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Nov 2020.

## Built on by (most cited first, 7 of 7)

- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [Sharing Lifelong Reinforcement Learning Knowledge via Modulating Masks](https://mlsanity.com/paper/2305.10997/) (2023)
- [Lightweight Learner for Shared Knowledge Lifelong Learning](https://mlsanity.com/paper/2305.15591/) (2023)
- [Class-Incremental Mixture of Gaussians for Deep Continual Learning](https://mlsanity.com/paper/2307.04094/) (2023)
- [CUAL: Continual Uncertainty-aware Active Learner](https://mlsanity.com/paper/2412.09701/) (2024)
- [Uncertainty Quantification in Continual Open-World Learning](https://mlsanity.com/paper/2412.16409/) (2024)
- [CONCLAD: COntinuous Novel CLAss Detector](https://mlsanity.com/paper/2412.10473/) (2024)

## Abstract (arXiv)

Supervised deep neural networks are known to undergo a sharp decline in the accuracy of older tasks when new tasks are learned, termed "catastrophic forgetting". Many state-of-the-art solutions to continual learning rely on biasing and/or partitioning a model to accommodate successive tasks incrementally. However, these methods largely depend on the availability of a task-oracle to confer task identities to each test sample, without which the models are entirely unable to perform. To address this shortcoming, we propose and compare several candidate task-assigning mappers which require very little memory overhead: (1) Incremental unsupervised prototype assignment using either nearest means, Gaussian Mixture Models or fuzzy ART backbones; (2) Supervised incremental prototype assignment with fast fuzzy ARTMAP; (3) Shallow perceptron trained via a dynamic coreset. Our proposed model variants are trained either from pre-trained feature extractors or task-dependent feature embeddings of the main classifier network. We apply these pipeline variants to continual learning benchmarks, comprised of either sequences of several datasets or within one single dataset. Overall, these methods, despite their simplicity and compactness, perform very close to a ground truth oracle, especially in experiments of inter-dataset task assignment. Moreover, best-performing variants only impose an average cost of 1.7% parameter memory increase.