Generative adversarial models are powerful tools to model structure in complex distributions for a variety of tasks. Current techniques for learning generative models require an access to samples which have high quality, and advanced generative models are applied to generate samples from noisy training data through ambient modules. However, the modules are only practical for the output space of the generator, and their application in the hidden space is not well studied. In this paper, we extend the ambient module to the hidden space of the generator, and provide the uniqueness condition and the corresponding strategy for the ambient hidden generator in the adversarial training process. We report the practicality of the proposed method on the benchmark dataset.

With the proliferation of social media, fashion inspired from celebrities, reputed designers as well as fashion influencers has shortened the cycle of fashion design and manufacturing. However, with the explosion of fashion related content and large number of user generated fashion photos, it is an arduous task for fashion designers to wade through social media photos and create a digest of trending fashion. This necessitates deep parsing of fashion photos on social media to localize and classify multiple fashion items from a given fashion photo. While object detection competitions such as MSCOCO have thousands of samples for each of the object categories, it is quite difficult to get large labeled datasets for fast fashion items. Moreover, state-of-the-art object detectors do not have any functionality to ingest large amount of unlabeled data available on social media in order to fine tune object detectors with labeled datasets. In this work, we show application of a generic object detector, that can be pretrained in an unsupervised manner, on 24 categories from recently released Open Images V4 dataset. We first train the base architecture of the object detector using unsupervisd learning on 60K unlabeled photos from 24 categories gathered from social media, and then subsequently fine tune it on 8.2K labeled photos from Open Images V4 dataset. On 300 X 300 image inputs, we achieve 72.7% mAP on a test dataset of 2.4K photos while performing 11% to 17% better as compared to the state-of-the-art object detectors. We show that this improvement is due to our choice of architecture that lets us do unsupervised learning and that performs significantly better in identifying small objects.

We propose a new regularization method based on virtual adversarial loss: a new measure of local smoothness of the conditional label distribution given input. Virtual adversarial loss is defined as the robustness of the conditional label distribution around each input data point against local perturbation. Unlike adversarial training, our method defines the adversarial direction without label information and is hence applicable to semi-supervised learning. Because the directions in which we smooth the model are only "virtually" adversarial, we call our method virtual adversarial training (VAT). The computational cost of VAT is relatively low. For neural networks, the approximated gradient of virtual adversarial loss can be computed with no more than two pairs of forward- and back-propagations. In our experiments, we applied VAT to supervised and semi-supervised learning tasks on multiple benchmark datasets. With a simple enhancement of the algorithm based on the entropy minimization principle, our VAT achieves state-of-the-art performance for semi-supervised learning tasks on SVHN and CIFAR-10.

The progress of technology in leaps and bounds has resulted in the generation of an enormous amount of digital data in the modern era. Against this backdrop, artificial intelligence (AI) has emerged as a useful mechanism to automatically organize and categorize data and to leverage useful patterns in the data to make intelligent predictions for the future from past observations. In this way, we can use AI to help older adults and others. The most common kind of AI algorithms is supervised machine learning, which involves learning from labeled data. However, while gathering a large amount of unlabeled data is cheap and easy, hand-labeling the data is an expensive process in terms of time, labor and human expertise.

Unsupervised learning sounds like a fancy way to say "let the kids learn on their own not to touch the hot oven", but it's actually a pattern-finding technique for mining inspiration from your data. Contrary to popular belief, it has nothing to do with machines running around without adult supervision, forming their own opinions about things. This post is beginner-friendly, but assumes you're familiar with the story so far: Check out these six instances. These photographs are just some pixel color data, but they're not accompanied by any labels. No worries, your human brain is pretty good at unsupervised learning.

Evaluating generative adversarial networks (GANs) is inherently challenging. In this paper, we revisit several representative sample-based evaluation metrics for GANs, and address the problem of how to evaluate the evaluation metrics. We start with a few necessary conditions for metrics to produce meaningful scores, such as distinguishing real from generated samples, identifying mode dropping and mode collapsing, and detecting overfitting. With a series of carefully designed experiments, we comprehensively investigate existing sample-based metrics and identify their strengths and limitations in practical settings. Based on these results, we observe that kernel Maximum Mean Discrepancy (MMD) and the 1-Nearest- Neighbor (1-NN) two-sample test seem to satisfy most of the desirable properties, provided that the distances between samples are computed in a suitable feature space. Our experiments also unveil interesting properties about the behavior of several popular GAN models, such as whether they are memorizing training samples, and how far they are from learning the target distribution.

Recent advances in deep unsupervised learning have renewed interest in semi-supervised methods, which can learn from both labeled and unlabeled data. Presently the most successful approaches to semi-supervised learning are based on consistency regularization, whereby a model is trained to be robust to small perturbations of its inputs and parameters. We show that consistency regularization leads to flatter but narrower optima. We also show that the test error surface for these methods is approximately convex in regions of weight space traversed by SGD. Inspired by these observations, we propose to train consistency based semi-supervised models with stochastic weight averaging (SWA), a recent method which averages weights along the trajectory of SGD. We also develop fast-SWA, which further accelerates convergence by averaging multiple points within each cycle of a cyclical learning rate schedule. With fast-SWA we achieve the best known semi-supervised results on CIFAR-10 and CIFAR-100 over many different numbers of observed training labels. For example, we achieve 95.0% accuracy on CIFAR-10 with only 4000 labels, compared to the previous best result in the literature of 93.7%. We also improve the best known accuracy for domain adaptation from CIFAR-10 to STL from 80% to 83%. Finally, we show that with fast-SWA the simple $\Pi$ model becomes state-of-the-art for large labeled settings.

June's NAACL conference saw machine learning specialists from technology company Iprova present a paper introducing a new and effective method for the unsupervised training of machine learning algorithms to infer sentence embeddings. The NAACL (North American Chapter of the Association for Computational Linguistics) Human Language Technologies (HLT) conference took place at the Hyatt Regency New Orleans hotel, Louisiana, from June 1–6, 2018. The research paper, entitled "Unsupervised Learning of Sentence Embeddings using Compositional n-Gram Features", will be presented by Matteo Pagliardini. Pagliardini is a senior machine learning engineer at Iprova and one of the three scientists that authored the research paper and developed the new model for unsupervised training, Sent2Vec. While there have been several successes in deep learning in recent years, the paper notes that these have almost exclusively relied on supervised training.

We present a novel cost function for semi-supervised learning of neural networks that encourages compact clustering of the latent space to facilitate separation. The key idea is to dynamically create a graph over embeddings of labeled and unlabeled samples of a training batch to capture underlying structure in feature space, and use label propagation to estimate its high and low density regions. We then devise a cost function based on Markov chains on the graph that regularizes the latent space to form a single compact cluster per class, while avoiding to disturb existing clusters during optimization. We evaluate our approach on three benchmarks and compare to state-of-the art with promising results. Our approach combines the benefits of graph-based regularization with efficient, inductive inference, does not require modifications to a network architecture, and can thus be easily applied to existing networks to enable an effective use of unlabeled data.

In this paper we develop Neural Random Field learning with Inclusive-divergence minimized Auxiliary Generators (NRF-IAG), which is under-appreciated in the literature. The contributions are two-fold. First, we rigorously apply the stochastic approximation algorithm to solve the joint optimization and provide theoretical justification. The new approach of learning NRF-IAG achieves superior unsupervised learning performance competitive with state-of-the-art deep generative models (DGMs) in terms of sample generation quality. Second, semi-supervised learning (SSL) with NRF-IAG gives rise to strong classification results comparable to state-of-art DGM-based SSL methods, and simultaneously achieves superior generation. This is in contrast to the conflict of good classification and good generation, as observed in GAN-based SSL.