On First-Order Meta-Learning Algorithms

8 Mar 2018Alex NicholJoshua AchiamJohn Schulman

This paper considers meta-learning problems, where there is a distribution of tasks, and we would like to obtain an agent that performs well (i.e., learns quickly) when presented with a previously unseen task sampled from this distribution. We analyze a family of algorithms for learning a parameter initialization that can be fine-tuned quickly on a new task, using only first-order derivatives for the meta-learning updates... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Few-Shot Image Classification Mini-Imagenet 5-way (1-shot) Reptile + Transduction Accuracy 49.97 # 39
Few-Shot Image Classification Mini-Imagenet 5-way (5-shot) Reptile + Transduction Accuracy 65.99 # 34
Few-Shot Image Classification OMNIGLOT - 1-Shot, 20-way Reptile + Transduction Accuracy 89.43% # 16
Few-Shot Image Classification OMNIGLOT - 1-Shot, 5-way Reptile + Transduction Accuracy 97.68 # 15
Few-Shot Image Classification OMNIGLOT - 5-Shot, 20-way Reptile + Transduction Accuracy 97.12% # 17
Few-Shot Image Classification OMNIGLOT - 5-Shot, 5-way Reptile + Transduction Accuracy 99.48 # 11

Results from Other Papers


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK SOURCE PAPER COMPARE
Image Classification Tiered ImageNet 5-way (5-shot) Reptile + BN Accuracy 71.03 # 3
Image Classification Tiered ImageNet 5-way (5-shot) Reptile Accuracy 66.47 # 7

Methods used in the Paper