Research

I am currently working on various topics related to machine learning, speech and language processing. If you are interested in doing research with me, please email me with your background.

Research Interests:

• Machine Learning / Pattern Classification
• Deep Learning / Neural Networks
• Natural Language Processing (language modelling, paraphrase, question and answering)
• Speech and Audio Processing (enhancement, recognition, human-machine dialogue)
• Computer Vision (object detection and cross-domain adaptation)

Currently Undergoing Research Projects:

• Neural Models for Natural Language Processing (Language Modelling, Paraphrase Detection, Q&A, Human-Machine Dialogue)
  1. Semantic Word Embedding
  2. Fixed-size Ordinally-Forgetting Encoding (FOFE)
  3. Neural Reading Machine

• Deep Learning for Computer Vision
  1. Deep Learning for Object Detection and Image Segmentation
  2. Neural Viewing Machine

• Deep Learning for Probabilistic Reasoning

• Learning Neural Networks under the HOPE Framework

• Feedforward Sequential Memory Network (FSMN)

Previous Research Projects:

• Distributed Training of Deep Neural Networks (DNNs) using GPUs and MapReduce

• Locally Connected Neural Networks

• Fast Stochastic Gradient Descent (SGD) Algorithm for Deep Learning

• Rapid Adaptation of Deep Neural Networks (DNNs)

• Multiple DNN Acoustic Models for Speech Recognition

• Convolutional Neural Networks (CNNs) for Speech Recognition

• Large Margin Estimation of HMMs for Speech Recognition

• Large Margin Multinomial Mixture Models for Text Categorization

• Noise Reduction and Speaker Normalization for Cochlear Implant Processing