Generative Latent Implicit Conditional Optimization when Learning from Small Sample

31 Mar 2020  ·  Idan Azuri, Daphna Weinshall ·

We revisit the long-standing problem of learning from a small sample, to which end we propose a novel method called GLICO (Generative Latent Implicit Conditional Optimization). GLICO learns a mapping from the training examples to a latent space and a generator that generates images from vectors in the latent space. Unlike most recent works, which rely on access to large amounts of unlabeled data, GLICO does not require access to any additional data other than the small set of labeled points. In fact, GLICO learns to synthesize completely new samples for every class using as little as 5 or 10 examples per class, with as few as 10 such classes without imposing any prior. GLICO is then used to augment the small training set while training a classifier on the small sample. To this end, our proposed method samples the learned latent space using spherical interpolation, and generates new examples using the trained generator. Empirical results show that the new sampled set is diverse enough, leading to improvement in image classification in comparison with the state of the art, when trained on small samples obtained from CIFAR-10, CIFAR-100, and CUB-200.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Small Data Image Classification CIFAR-100, 1000 Labels GLICO Accuracy 28.55 # 3
Small Data Image Classification CIFAR-10, 250 Labels GLICO Top-1 accuracy % 43 # 1
Small Data Image Classification CIFAR-10, 500 Labels GLICO Accuracy (%) 56.22 # 6
Small Data Image Classification CUB-200-2011, 30 samples per class GLICO Accuracy 77.75 # 1
Small Data Image Classification CUB-200-2011, 5 samples per class GLICO Accuracy 51.52 # 1