• Publications
  • Influence
A Mixture of Experts Classifier with Learning Based on Both Labelled and Unlabelled Data
TLDR
We address classifier design given a mixed training set consisting of a small labelled feature set and a (generally larger) set of unlabelled features. Expand
  • 332
  • 16
  • PDF
Backdoor Embedding in Convolutional Neural Network Models via Invisible Perturbation
TLDR
Deep learning models have consistently outperformed traditional machine learning models in various classification tasks, including image classification. Expand
  • 72
  • 15
  • PDF
A sequence-based approximate MMSE decoder for source coding over noisy channels using discrete hidden Markov models
TLDR
In previous work on source coding over noisy channels it was recognized that when the source has memory, there is typically "residual redundancy" between the discrete symbols produced by the encoder, which can be capitalized upon by the decoder to improve the overall quantizer performance. Expand
  • 73
  • 8
Joint source-channel decoding for variable-length encoded data by exact and approximate MAP sequence estimation
TLDR
We describe and compare the performance of a computationally complex exact maximum a posteriori (MAP) decoder, its efficient approximation, and an alternative approximate decoder for variable-length encoded data. Expand
  • 92
  • 7
Content-Driven Detection of Cyberbullying on the Instagram Social Network
TLDR
We study detection of cyberbullying in photo-sharing networks, with an eye on developing early-warning mechanisms for the prediction of posted images vulnerable to attacks. Expand
  • 80
  • 7
  • PDF
Mixture of experts regression modeling by deterministic annealing
We propose a new learning algorithm for regression modeling. The method is especially suitable for optimizing neural network structures that are amenable to a statistical description as mixtureExpand
  • 70
  • 7
  • PDF
Joint source-channel decoding for variable-length encoded data by exact and approximate MAP sequence estimation
TLDR
Joint source-channel decoding based on residual source redundancy is an effective paradigm for error-resilient data compression and provides substantial error protection to variable-length encoded data. Expand
  • 60
  • 7
Parsimonious Topic Models with Salient Word Discovery
TLDR
We propose a parsimonious topic model for text corpora that gives sparse topic representation, determining the (small) subset of relevant topics for each document. Expand
  • 26
  • 6
  • PDF
MDA5 and TLR3 Initiate Pro-Inflammatory Signaling Pathways Leading to Rhinovirus-Induced Airways Inflammation and Hyperresponsiveness
Rhinovirus (RV), a single-stranded RNA picornavirus, is the most frequent cause of asthma exacerbations. We previously demonstrated in human bronchial epithelial cells that melanomaExpand
  • 110
  • 5
  • PDF
A Deterministic Annealing Approach for Parsimonious Design of Piecewise Regression Models
TLDR
A new learning algorithm is proposed for piecewise regression modeling. Expand
  • 72
  • 5
...
1
2
3
4
5
...