June 23rd(Sun) - 26th(Wed), 2019/ Jeju Shinhwa World, Republic of Korea
Submission of paper
March 22, 2019March 28, 2019
Notification of Acceptance
April 19, 2019
Submission of Final Paper
May 10, 2019
TODAY 2019. 03. 24
Lightweight Benchmark Suite for Neural Networks
William J. Song, Ph.D.
- Yonsei University, Korea
This tutorial presents a lightweight benchmark suite for neural networks. The advance of computing systems and explosive growth of data production have sparked the unprecedentedly rapid evolution of machine learning. Neural networks and deep learning provide state-of-the-art algorithms designed to recognize patterns. Neural network structures resemble human brains by organizing data into neurons and synapses. The computation of neural networks under the hood is represented as repeated matrix or vector arithmetic on conventional Von Neumann architectures. Recent neural networks tend to form deep networks by increasing the volume of layers and dataset to enhance accuracy. However, such a trend imposes modeling and validation challenges on the analysis and development of accelerator hardware, since it requires increasingly longer execution time to process the sizable data and operation count of a neural network. In this tutorial, we present and demonstrate a novel lightweight benchmark suite for neural networks to tackle the aforementioned engineering challenges.
William J. Song is currently an Assistant Professor with the School of Electrical and Electronic Engineering, Yonsei University in Seoul, South Korea. He earned his Ph.D. degree in Electrical and Computer Engineering from Georgia Tech, Atlanta, GA, and B.S. degree in Electrical and Electronic Engineering from Yonsei University, Seoul, South Korea. His research focus lies in the challenges of heterogeneous architectures and processing near data for neural networks and big data problems. His interests also include solutions to power, thermal, and reliability issues in many-core microarchitectures and 3D-integrated packages. Prior to joining the faculty of Yonsei University, he was a senior engineer at Intel in Santa Clara, CA. He was a graduate research intern at Qualcomm, San Diego, CA (2015 summer), IBM T.J. Watson Research Center, Yorktown Heights, NY (2014 summer and fall), AMD Research, Bellevue, WA (2013 summer), and Sandia National Labs, Albuquerque, NM (2012, 2011, and 2010 summers). He received Distinguished Faculty Award for Teaching Excellence from Yonsei University in 2018. He was a recipient of IBM/SRC graduate fellowship from 2012 to 2015. He received the Best Student Paper Award at IEEE International Reliability Physics Symposium (IRPS) in 2015 and Best in Session Award at SRC TECHCON in 2014.
Neural network and its variants
Prof. Shun Kataoka
- Otaru University of Commerce, Japan
Recent developments of machine learning bring innovations to various scientific areas. Especially, the developments of the computer vision area brought by the convolutional neural networks is remarkable and neural networks are becoming one of the basic techniques for computer vision. Neural network itself is a classic machine learning model and it is not difficult to understand the its basic property. In this talk, I will explain the foundation of the neural network with its approximation property. Then I will survey variants of the neural network including convolutional neural network and recurrent neural network.
Shun Kataoka is an associate professor at Otaru University of Commerce(OUC). He received the doctorial degree (information science) from Tohoku University in 2014. He worked at Tohoku University as an assistant professor for 4 years before moving to OUC in 2018. His research interests are statistical machine learning, computer vision, complex networks, and statistical mechanics.
Artificial Intelligence for Medical Screening and Diagnosis: Research and Innovation
- Thammasat University, Thailand
Artificial Intelligence (AI) has been applied to a lot of medical applications especially for disease screening and diagnosis. One objective is to support medical staffs especially in developing countries, where medical experts and resources are very limited. As a result, patients can have medical screening and can receive medical treatment on time reducing disability and loss of life. Our research group, emphasized in AI in medicine, have collaborated on interdisciplinary research and development of medical innovations using resources available in Thailand. Our ultimate goal is to facilitate physicians, public health officers, and people in medical screening and diagnosis, focusing on the major diseases, e.g., stroke, Alzheimer’s, learning disability, diabetic retinopathy, aged-macular degeneration, glaucoma, cytomegalovirus retinitis, cervical cancer, skin cancer, lung cancer, and tuberculosis that are affecting majority people around the world. Based on our continuous dedication for more than 10 years, we have developed a lot of innovative medical products with world class quality. These innovations have high impacts resulting in the better quality of life of Thai people. Our research work and innovations have been perceived as one of the best research groups in Thailand as shown by awards received nationally and internationally. Moreover, our research group is only one that won the Grand Prize in International Exhibition of Inventions of Geneva, Switzerland recognized as the world’s largest exhibition of inventions. Finally, our research work and innovations have been published in the world leading international journals.
CHARTURONG TANTIBUNDHIT received the B.E. degree in electrical engineering from Kasetsart University, Bangkok, Thailand, in 1996, and the M.S. degree in information science and Ph.D. degree in electrical engineering from the University of Pittsburgh, Pittsburgh, PA, USA, in 2001 and 2006, respectively. Since 2006, he has been with Thammasat University, Thailand, where he is currently an Associate Professor with the Department of Electrical and Computer Engineering and the Head of the Speech and Language Technology Cluster, Center of Excellence in Intelligence Informatics, Speech and Language Technology, and Service Innovation. From 2007 to 2008, he was a Post-Doctoral Researcher with the Signal Processing and Speech Communication Laboratory, Graz University of Technology, Graz, Austria. He was an IEEE ICASSP Student Paper Contest Winner in 2006. He led a team to win the Grand Prix of the 45th International Exhibition of Inventions of Geneva in 2017. His research interests include handcrafted machine learning and deep learning in medicine, biomedical signal processing, and speech processing.