Few-shot learning aims to classify unseen classes with a few training examples. While recent works have shown that standard mini-batch training with a carefully designed training strategy can improve generalization ability for unseen classes, well-known problems in deep networks such as memorizing training statistics have been less explored for few-shot learning. To tackle this issue, we propose self-augmentation that consolidates regional dropout and self-distillation. Specifically, we exploit a data augmentation technique called regional dropout, in which a patch of an image is substituted into other values. Then, we employ a backbone network that has auxiliary branches with its own classifier to enforce knowledge sharing. Lastly, we present a fine-tuning method to further exploit a few training examples for unseen classes. Experimental results show that the proposed method outperforms the state-of-the-art methods for prevalent few-shot benchmarks and improves the generalization ability.

Yann Lecun, in his talk, introduced the "cake analogy" to illustrate the importance of self-supervised learning. Though the analogy is debated(ref: Deep Learning for Robotics(Slide 96), Pieter Abbeel), we have seen the impact of self-supervised learning in the Natural Language Processing field where recent developments (Word2Vec, Glove, ELMO, BERT) have embraced self-supervision and achieved state of the art results. "If intelligence is a cake, the bulk of the cake is self-supervised learning, the icing on the cake is supervised learning, and the cherry on the cake is reinforcement learning (RL)." Curious to know how self-supervised learning has been applied in the computer vision field, I read up on existing literature on self-supervised learning applied to computer vision through a recent survey paper by Jing et. This post is my attempt to provide an intuitive visual summary of the patterns of problem formulation in self-supervised learning.

In this work, we focus on the use of influence functions to identify relevant training examples that one might hope "explain" the predictions of a machine learning model. One shortcoming of influence functions is that the training examples deemed most "influential" are often outliers or mislabelled, making them poor choices for explanation. In order to address this shortcoming, we separate the role of global versus local influence. We introduce RelatIF, a new class of criteria for choosing relevant training examples by way of an optimization objective that places a constraint on global influence. RelatIF considers the local influence that an explanatory example has on a prediction relative to its global effects on the model. In empirical evaluations, we find that the examples returned by RelatIF are more intuitive when compared to those found using influence functions.

Many training algorithms of a deep neural network can be interpreted as minimizing the cross entropy loss between the prediction made by the network and a target distribution. In supervised learning, this target distribution is typically the ground-truth one-hot vector. In semi-supervised learning, this target distribution is typically generated by a pre-trained teacher model to train the main network. In this work, instead of using such predefined target distributions, we show that learning to adjust the target distribution based on the learning state of the main network can lead to better performances. In particular, we propose an efficient meta-learning algorithm, which encourages the teacher to adjust the target distributions of training examples in the manner that improves the learning of the main network. The teacher is updated by policy gradients computed by evaluating the main network on a held-out validation set. Our experiments demonstrate substantial improvements over strong baselines and establish state-ofthe-art performance on CIFAR-10, SVHN, and ImageNet. For instance, with ResNets on small datasets, we achieve 96.1% on CIFAR-10 with 4,000 labeled examples and 73.9% top-1 on ImageNet with 10% examples. Meanwhile, with EfficientNet on full datasets plus extra unlabeled data, we attain 98.6% accuracy on CIFAR-10 and 86.9% top-1 accuracy on ImageNet.

So how can we improve the model? One approach is to continue to train our model on our image set but during the training we will generate adversarial noise that we add to the image. Since we're training our model, we still know all the labels of our images and we can train the model to classify the images according to the specific label even when the image contains particular noise. This method of'adversarial training' helps generalize the model and makes it more robust against noise that the images might include. It therefore makes the model less likely to make wrong predictions when images outside the training set contain perturbations.

Machine teaching studies the interaction between a teacher and a student/learner where the teacher selects training examples for the learner to learn a specific task. The typical assumption is that the teacher has perfect knowledge of the task---this knowledge comprises knowing the desired learning target, having the exact task representation used by the learner, and knowing the parameters capturing the learning dynamics of the learner. Inspired by real-world applications of machine teaching in education, we consider the setting where teacher's knowledge is limited and noisy, and the key research question we study is the following: When does a teacher succeed or fail in effectively teaching a learner using its imperfect knowledge? We answer this question by showing connections to how imperfect knowledge affects the teacher's solution of the corresponding machine teaching problem when constructing optimal teaching sets. Our results have important implications for designing robust teaching algorithms for real-world applications.

We address the problem of general supervised learning when data can only be accessed through an (indefinite) similarity function between data points. Existing work on learning with indefinite kernels has concentrated solely on binary/multiclass classification problems. We propose a model that is generic enough to handle any supervised learning task and also subsumes the model previously proposed for classification. We give a ''goodness'' criterion for similarity functions w.r.t. a given supervised learning task and then adapt a well-known landmarking technique to provide efficient algorithms for supervised learning using ''good'' similarity functions. We demonstrate the effectiveness of our model on three important supervised learning problems: a) real-valued regression, b) ordinal regression and c) ranking where we show that our method guarantees bounded generalization error.

Deep structured output learning shows great promise in tasks like semantic image segmentation. We proffer a new, efficient deep structured model learning scheme, in which we show how deep Convolutional Neural Networks (CNNs) can be used to directly estimate the messages in message passing inference for structured prediction with Conditional Random Fields CRFs). With such CNN message estimators, we obviate the need to learn or evaluate potential functions for message calculation. This confers significant efficiency for learning, since otherwise when performing structured learning for a CRF with CNN potentials it is necessary to undertake expensive inference for every stochastic gradient iteration. The network output dimension of message estimators is the same as the number of classes, rather than exponentially growing in the order of the potentials.

Self-supervision provides effective representations for downstream tasks without requiring labels. However, existing approaches lag behind fully supervised training and are often not thought beneficial beyond obviating or reducing the need for annotations. We find that self-supervision can benefit robustness in a variety of ways, including robustness to adversarial examples, label corruption, and common input corruptions. Additionally, self-supervision greatly benefits out-of-distribution detection on difficult, near-distribution outliers, so much so that it exceeds the performance of fully supervised methods. These results demonstrate the promise of self-supervision for improving robustness and uncertainty estimation and establish these tasks as new axes of evaluation for future self-supervised learning research. Papers published at the Neural Information Processing Systems Conference.

We study the task of semi-supervised learning on multilayer graphs by taking into account both labeled and unlabeled observations together with the information encoded by each individual graph layer. We propose a regularizer based on the generalized matrix mean, which is a one-parameter family of matrix means that includes the arithmetic, geometric and harmonic means as particular cases. We analyze it in expectation under a Multilayer Stochastic Block Model and verify numerically that it outperforms state of the art methods. Moreover, we introduce a matrix-free numerical scheme based on contour integral quadratures and Krylov subspace solvers that scales to large sparse multilayer graphs. Papers published at the Neural Information Processing Systems Conference.