MearcAll papers

arXiv 1703.03400 · cs.LG cs.AI cs.CV

Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks

Chelsea Finn, Pieter Abbeel, Sergey Levine

Claims were extracted automatically by reducer-1 on 2026-11-12. The paper's license does not allow republishing it, so long claims are shortened to a quotation. Read the full text on arXiv.

No reproduction recordedclm_4102e6bc
We propose an algorithm for meta-learning that is model-agnostic, in the sense that it is compatible with any model trained with gradient descent and applicable to a variety of different … (excerpt)

Abstract · page 1 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_507eeee1
We demonstrate that this approach leads to state-of-the-art performance on two fewshot image classification benchmarks, produces good results on few-shot regression, and accelerates fine-tuning for policy gradient reinforcement learning with … (excerpt)

Abstract · page 1 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_6308b866
In this work, we propose a meta-learning algorithm that is general and model-agnostic, in the sense that it can be directly applied to any learning problem and model that is … (excerpt)

Introduction · page 1 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_f5640da1
From a dynamical systems standpoint, our learning process can be viewed as maximizing the sensitivity of the loss functions of new tasks with respect to the parameters: when the sensitivity … (excerpt)

Introduction · page 1 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_42d21d2b
We demonstrate the algorithm on different model types, including fully connected and convolutional networks, and in several distinct domains, including few-shot regression, image classification, and reinforcement learning.

Introduction · page 2 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_dfad2922
Our evaluation shows that our meta-learning algorithm compares favorably to state-of-the-art one-shot learning methods designed specifically for supervised classification, while using fewer parameters, but that it can also be readily … (excerpt)

Introduction · page 2 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_c714a628
In effect, we will aim to find model parameters that are sensitive to changes in the task, such that small changes in the parameters will produce large improvements on the … (excerpt)

A Model-Agnostic Meta-Learning Algorithm · page 2 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_d1d0d4fd
(3) Can a model learned with MAML continue to improve with additional gradient updates and/or examples?

Experimental Evaluation · page 5 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_9e562dce
Furthermore, we observe both in the qualitative and quantitative results (Figure 3 and Appendix B) that the model learned with MAML continues to improve with additional gradient steps, despite being … (excerpt)

Regression · page 6 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_47063ac0
This improvement suggests that MAML optimizes the parameters such that they lie in a region that is amenable to fast adaptation and is sensitive to loss functions fromp(T ), as … (excerpt)

Regression · page 6 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_87572713
MAML achieves results that are comparable to or outperform state-of-the-art convolutional and recurrent models.

Regression · page 7 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_d3b15eb0
However, as shown in the comparison, MAML significantly outperforms memory-augmented networks and the meta-learner LSTM on 5-way Omniglot and MiniImagenet classification, both in the 1-shot and 5-shot case.

Regression · page 7 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_47d6892d
The results in Figure 5 show that MAML learns a model that can quickly adapt its velocity and direction with even just a single gradient update, and continues to improve … (excerpt)

Reinforcement Learning · page 8 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_a74949f5
The results also show that, on these challenging tasks, the MAML initialization substantially outperforms random initialization and pretraining.

Reinforcement Learning · page 8 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_c60c834c
Lastly, since our method merely produces a weight initialization, adaptation can be performed with any amount of data and any number of gradient steps, though we demonstrate state-of-the-art results on … (excerpt)

Discussion and Future Work · page 8 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_feeb11ab
To achieve averaging in parameter space, we sequentially trained 500 separate models on 500 tasks drawn from p(T ).

Multi-task baselines · page 11 · Automated extraction, not reviewed · 0 reviewed reproductions

Reproduced one of these claims?

Record your result in the Mearc app on your own machine, then use Publish this result with this paper's id (1703.03400) and the claim id shown next to the claim. Mearc creates a file that holds only what you choose to publish. Add it to the public claim data by pull request; a maintainer other than you reviews it before it counts.