Abstract: Many popular machine learning models for prediction and classification can be described as a series of computation steps. Such models can be represented using a structure known as a computational network. A computational network expresses a model’s operation as a graph, where leaf nodes represent input values or learnable parameters and parent nodes represent basic computations, such as sum, multiplication, or logarithm. Arbitrarily complex computations can be performed using a sequence of such nodes. Deep neural networks, convolutional neural networks, recurrent neural networks and maximum entropy models are all examples of models that can be expressed using computational networks.

In this talk I will first introduce computational networks (CNs) and describe the benefits of such generalization and the key algorithms involved in CNs. I will then introduce the computational network toolkit (CNTK), a general purpose C++ implementation of computational networks. I will describe its architecture and core functionalities and demonstrate that it can construct and learn models of arbitrary topology, connectivity, and recurrence.

Biography: Dr. Dong Yu is a principal researcher at the Microsoft speech and dialog research group. His current research interests include speech processing, robust speech recognition, discriminative training, and machine learning. He has published over 140 papers in these areas and is the co-inventor of more than 50 granted/pending patents. His work on context-dependent deep neural network hidden Markov model (CD-DNN-HMM) has helped to shape the new direction on large vocabulary speech recognition research and was recognized by the IEEE SPS 2013 best paper award. Most recently he has focused on applying computational networks, a generalization of many neural network models, to speech recognition.

Dr. Dong Yu is currently serving as a member of the IEEE Speech and Language Processing Technical Committee (2013-) and an associate editor of IEEE transactions on audio, speech, and language processing (2011-). He has served as an associate editor of IEEE signal processing magazine (2008-2011) and the lead guest editor of IEEE transactions on audio, speech, and language processing - special issue on deep learning for speech and language processing (2010-2011).

contact: Prof. Jia Liu, liuj@tsinghua.edu.cn  Dept. of E.E., Tsinghua University