• Publications
  • Influence
Co-regularized Multi-view Spectral Clustering
TLDR
A spectral clustering framework is proposed that achieves this goal by co-regularizing the clustering hypotheses, and two co- regularization schemes are proposed to accomplish this.
Generalized Zero-Shot Learning via Synthesized Examples
TLDR
This work presents a generative framework for generalized zero-shot learning where the training and test classes are not necessarily disjoint, and can generate novel exemplars from seen/unseen classes, given their respective class attributes.
A Deep Generative Framework for Paraphrase Generation
TLDR
Quantitative evaluation of the proposed method on a benchmark paraphrase dataset demonstrates its efficacy, and its performance improvement over the state-of-the-art methods by a significant margin, whereas qualitative human evaluation indicate that the generated paraphrases are well-formed, grammatically correct, and are relevant to the input sentence.
Online Learning of Multiple Tasks and Their Relationships
TLDR
This work proposes an Online MultiTask Learning (Omtl) framework which simultaneously learns the task weight vectors as well as the task relatedness adaptively from the data, and exploits this adaptively learned task-relationship matrix to select the most informative samples in an online multitask active learning setting.
Scalable Bayesian Low-Rank Decomposition of Incomplete Multiway Tensors
TLDR
A scalable Bayesian framework for low-rank decomposition of multiway tensor data with missing observations, which outperforms several state-of-the-art tensor decomposition methods on various synthetic and benchmark real-world datasets.
Multiview Clustering with Incomplete Views
TLDR
This work presents an approach that allows the multiview clustering algorithms to be applicable even when only one (the primary) view is complete and the auxiliary views are incomplete (i.e., features from these views are av ailable only for some of the examples).
Zero-Shot Learning via Class-Conditioned Deep Generative Models
TLDR
A deep generative model for Zero-Shot Learning that represents each seen/unseen class using a class-specific latent-space distribution, conditioned on class attributes, which facilitates learning highly discriminative feature representations for the inputs.
A Simple Exponential Family Framework for Zero-Shot Learning
TLDR
A simple generative framework for learning to predict previously unseen classes, based on estimating class-attribute-gated class-conditional distributions, which extends seamlessly to few-shot learning by easily updating these distributions when provided with a small number of additional labelled examples from unseen classes.
Simultaneously Leveraging Output and Task Structures for Multiple-Output Regression
TLDR
This paper presents a multiple-output regression model that leverages the covariance structure of the latent model parameters as well as the conditional covarianceructure of the observed outputs, in contrast with existing methods that usually take into account only one of these structures.
A Generative Approach to Zero-Shot and Few-Shot Action Recognition
We present a generative framework for zero-shot action recognition where some of the possible action classes do not occur in the training data. Our approach is based on modeling each action class
...
...