Don’t Wait, Just Weight: Improving Unsupervised Representations by Learning Goal-Driven Instance Weights

22 Jun 2020  ·  Linus Ericsson ·

In the absence of large labelled datasets, self-supervised learning techniques can boost performance by learning useful representations from unlabelled data, which is often more readily available. However, there is often a domain shift between the unlabelled collection and the downstream target problem data. We show that by learning Bayesian instance weights for the unlabelled data, we can improve the downstream classification accuracy by prioritising the most useful instances. Additionally, we show that the training time can be reduced by discarding unnecessary datapoints. Our method, BetaDataWeighter is evaluated using the popular self-supervised rotation prediction task on STL-10 and Visual Decathlon. We compare to related instance weighting schemes, both hand-designed heuristics and meta-learning, as well as conventional self-supervised learning. BetaDataWeighter achieves both the highest average accuracy and rank across datasets, and on STL-10 it prunes up to 78% of unlabelled images without significant loss in accuracy, corresponding to over 50% reduction in training time.



Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Image Classification STL-10 BDW Percentage correct 71.12 # 88
Image Classification STL-10 L2RW Percentage correct 63.13 # 100
Image Classification STL-10 NN-Weighter Percentage correct 69.15 # 94
Image Classification STL-10 RotNet Percentage correct 68.19 # 97


No methods listed for this paper. Add relevant methods here