Conditional random fields (CRFs) are commonly employed as a post-processing tool for image segmentation tasks. The unary potentials of the CRF are often learnt independently by a classifier, thereby decoupling the inference in CRF from the training of classifier. Such a scheme works effectively, when pixel-level labelling is available for all the images. However, in absence of pixel-level labels, the classifier is faced with the uphill task of selectively assigning the image-level labels to the pixels of the image. Prior work often relied on localization cues, such as saliency maps, objectness priors, bounding boxes etc., to address this challenging problem. In contrast, we model the labels of the pixels as latent variables of a CRF. The pixels and the image-level labels are the observed variables of the latent CRF. We amortize the cost of inference in the latent CRF over the entire dataset, by training an inference network to approximate the posterior distribution of the latent variables given the observed variables. The inference network can be trained in an end-to-end fashion, and requires no localization cues for training. Moreover, unlike other approaches for weakly-supervised segmentation, the proposed model doesn't require further post-processing. The proposed model achieves performance comparable with other approaches that employ saliency masks for the task of weakly-supervised semantic image segmentation on the challenging VOC 2012 dataset.
Deep neural networks achieve unprecedented performance levels over many tasks and scale well with large quantities of data, but performance in the low-data regime and tasks like one shot learning still lags behind. While recent work suggests many hypotheses from better optimization to more complicated network structures, in this work we hypothesize that having a learnable and more expressive similarity objective is an essential missing component. Towards overcoming that, we propose a network design inspired by deep residual networks that allows the efficient computation of this more expressive pairwise similarity objective. Further, we argue that regularization is key in learning with small amounts of data, and propose an additional generator network based on the Generative Adversarial Networks where the discriminator is our residual pairwise network. This provides a strong regularizer by leveraging the generated data samples. The proposed model can generate plausible variations of exemplars over unseen classes and outperforms strong discriminative baselines for few shot classification tasks. Notably, our residual pairwise network design outperforms previous state-of-theart on the challenging mini-Imagenet dataset for one shot learning by getting over 55% accuracy for the 5-way classification task over unseen classes.
We propose a neural network based approach for learning topics from text and image datasets. The model makes no assumptions about the conditional distribution of the observed features given the latent topics. This allows us to perform topic modelling efficiently using sentences of documents and patches of images as observed features, rather than limiting ourselves to words. Moreover, the proposed approach is online, and hence can be used for streaming data. Furthermore, since the approach utilizes neural networks, it can be implemented on GPU with ease, and hence it is very scalable.
In this work we use the recent advances in representation learning to propose a neural architecture for the problem of natural language inference. Our approach is aligned to mimic how a human does the natural language inference process given two statements. The model uses variants of Long Short Term Memory (LSTM), attention mechanism and composable neural networks, to carry out the task. Each part of our model can be mapped to a clear functionality humans do for carrying out the overall task of natural language inference. The model is end-to-end differentiable enabling training by stochastic gradient descent. On Stanford Natural Language Inference(SNLI) dataset, the proposed model achieves better accuracy numbers than all published models in literature.
Recently there has been an enormous interest in generative models for images in deep learning. In pursuit of this, Generative Adversarial Networks (GAN) and Variational Auto-Encoder (VAE) have surfaced as two most prominent and popular models. While VAEs tend to produce excellent reconstructions but blurry samples, GANs generate sharp but slightly distorted images. In this paper we propose a new model called Variational InfoGAN (ViGAN). Our aim is two fold: (i) To generated new images conditioned on visual descriptions, and (ii) modify the image, by fixing the latent representation of image and varying the visual description. We evaluate our model on Labeled Faces in the Wild (LFW), celebA and a modified version of MNIST datasets and demonstrate the ability of our model to generate new images as well as to modify a given image by changing attributes.
Variational autoencoders (VAEs), that are built upon deep neural networks have emerged as popular generative models in computer vision. Most of the work towards improving variational autoencoders has focused mainly on making the approximations to the posterior flexible and accurate, leading to tremendous progress. However, there have been limited efforts to replace pixel-wise reconstruction, which have known shortcomings. In this work, we use real-valued non-volume preserving transformations (real NVP) to exactly compute the conditional likelihood of the data given the latent distribution. We show that a simple VAE with this form of reconstruction is competitive with complicated VAE structures, on image modeling tasks. As part of our model, we develop powerful conditional coupling layers that enable real NVP to learn with fewer intermediate layers.
In this paper, we address the problem of conditional modality learning, whereby one is interested in generating one modality given the other. While it is straightforward to learn a joint distribution over multiple modalities using a deep multimodal architecture, we observe that such models aren't very effective at conditional generation. Hence, we address the problem by learning conditional distributions between the modalities. We use variational methods for maximizing the corresponding conditional log-likelihood. The resultant deep model, which we refer to as conditional multimodal autoencoder (CMMA), forces the latent representation obtained from a single modality alone to be `close' to the joint representation obtained from multiple modalities. We use the proposed model to generate faces from attributes. We show that the faces generated from attributes using the proposed model, are qualitatively and quantitatively more representative of the attributes from which they were generated, than those obtained by other deep generative models. We also propose a secondary task, whereby the existing faces are modified by modifying the corresponding attributes. We observe that the modifications in face introduced by the proposed model are representative of the corresponding modifications in attributes.
The aim of the paper is to provide an exact approach for generating a Poisson process sampled from a hierarchical CRM, without having to instantiate the infinitely many atoms of the random measures. We use completely random measures~(CRM) and hierarchical CRM to define a prior for Poisson processes. We derive the marginal distribution of the resultant point process, when the underlying CRM is marginalized out. Using well known properties unique to Poisson processes, we were able to derive an exact approach for instantiating a Poisson process with a hierarchical CRM prior. Furthermore, we derive Gibbs sampling strategies for hierarchical CRM models based on Chinese restaurant franchise sampling scheme. As an example, we present the sum of generalized gamma process (SGGP), and show its application in topic-modelling. We show that one can determine the power-law behaviour of the topics and words in a Bayesian fashion, by defining a prior on the parameters of SGGP.
Hypergraph partitioning lies at the heart of a number of problems in machine learning and network sciences. Many algorithms for hypergraph partitioning have been proposed that extend standard approaches for graph partitioning to the case of hypergraphs. However, theoretical aspects of such methods have seldom received attention in the literature as compared to the extensive studies on the guarantees of graph partitioning. For instance, consistency results of spectral graph partitioning under the stochastic block model are well known. In this paper, we present a planted partition model for sparse random non-uniform hypergraphs that generalizes the stochastic block model. We derive an error bound for a spectral hypergraph partitioning algorithm under this model using matrix concentration inequalities. To the best of our knowledge, this is the first consistency result related to partitioning non-uniform hypergraphs.
Smoothed functional (SF) schemes for gradient estimation are known to be efficient in stochastic optimization algorithms, specially when the objective is to improve the performance of a stochastic system. However, the performance of these methods depends on several parameters, such as the choice of a suitable smoothing kernel. Different kernels have been studied in literature, which include Gaussian, Cauchy and uniform distributions among others. This paper studies a new class of kernels based on the q-Gaussian distribution, that has gained popularity in statistical physics over the last decade. Though the importance of this family of distributions is attributed to its ability to generalize the Gaussian distribution, we observe that this class encompasses almost all existing smoothing kernels. This motivates us to study SF schemes for gradient estimation using the q-Gaussian distribution. Using the derived gradient estimates, we propose two-timescale algorithms for optimization of a stochastic objective function in a constrained setting with projected gradient search approach. We prove the convergence of our algorithms to the set of stationary points of an associated ODE. We also demonstrate their performance numerically through simulations on a queuing model.