Loading Events
  • This event has passed.

a2-dlearn: The Ann Arbor Deep Learning Event

November 7, 2015 @ 1:30 pm - 7:30 pm

We are pleased to announce the first Ann Arbor Deep Learning event — a joint production of the Ann Arbor – Natural Language Processing and the Machine Learning – Data, Science and Industry meetups.  This mini-conference/workshop will feature speakers working in deep learning from a variety of backgrounds, and is open to anyone interested in deep learning concepts, implementation and implications across various fields. Registration  A2 Deep Learning Event Speakers Xin Rong is a PhD candidate at the University of Michigan School of Information. He works with Prof. Eytan Adar on text mining, natural language processing, and social network analysis. His research is focused on computational modeling of human communication behavior using advanced language models and data mining techniques. His research contributes to better understanding and prediction of human communication behavior in various contexts, including marketing, problem solving, and troubleshooting. He interned twice as a software engineer at Google working on projects related to language models and word embedding techniques. Catherine Finegan-Dollak is a PhD student at the University of Michigan, studying natural language processing (NLP) as part of the CLAIR group, supervised by Professor Dragomir Radev.  She received her bachelor’s degree from Boston College and her juris doctorate from the University of Virginia School of Law. She practiced law for several years. She is interested in semantics: What information is in this document, how can we represent it, and what can we do with that representation?  Currently she is exploring how semantics can be used for automatic summarization and is interested in how deep learning can be applied to NLP problems. Daniel Pressel is Chief Science Officer at Digital Roots, working on NLP and Machine Learning.  Prior to Digital Roots, Daniel was at General Dynamics – Advanced Information Systems, working as a technical lead and principal investigator on various large software projects that pushed the bounds of technology in the field of Image Processing.  He is obsessed with Machine Learning, NLP and Information Retrieval (IR) and is the organizer of the Ann Arbor/Detroit NLP Meetup, and co-organizer of the Machine Learning – Data, Science and Industry meetup. Micah Bojrab is a seasoned software engineer at MDA and a PhD student under Ming Dong at Wayne State University working on GPU-based Convolutional Neural Nets for Image Recognition.   Micah has an extensive commercial background working on parallelization of algorithms using CUDA. Daniel Zhou received a PhD in a Machine Learning-related field from the University of Michigan School of Information in 2013.  He is now a data science consultant and an entrepreneur.  Previously he worked for IBM as a software engineer.  Daniel is a founder and co-organizer of the Machine Learning – Data, Science and Industry meetup in Ann Arbor. Gint Puskorius is Manager and Senior Technical Leader for Ford’s Speech and Signal Processing Research and Advanced Engineering organization.  Gint joined Ford Motor Company’s research organization in 1982, after graduating from John Carroll University with  MS in Physics.  His early research at Ford focused on the then emerging fields of computer vision and robotics, with a focus on manufacturing applications.  In the late 1980’s, Gint became a charter member of Ford’s Artificial Neural Networks group.  In this work, Gint was responsible for development of Kalman filter-based learning algorithms with application to recurrent neural networks for problems in diagnostics and controls.  Some of this work was recognized by a 1995 IEEE Transactions on Neural Networks Outstanding Paper Award.  In 1997, Gint helped to form a Business Analytics organization within Ford’s Research Lab.  This activity has grown over the years, has been recently recognized externally for contributions to Ford’s business, and is now part of Ford’s Global Data Insights and Analytics organization.  In 2013, Gint assumed responsibilities for growing Ford’s competencies in the areas of speech recognition and signal processing.  While starting that activity, Gint also led Ford’s efforts throughout 2104 in the planning, design, build-out and preliminary staffing of Ford’s Research and Innovation Center – Palo Alto, which was opened in January 2015. Honglak Lee is a recognized authority in the field of deep learning. His research interests lie in machine learning and its applications to artificial intelligence. In particular, his focus is on deep learning and representation learning, which aims to learn an abstract representation of the data by a hierarchical and compositional structure. His research also spans over related topics, such as graphical models, optimization, and large-scale learning. Specific application areas include computer vision, audio recognition, robotics, text modeling, and healthcare.

Schedule 1:30 – 2:15 Micah BojrabThinking Neural Networks 2:30 – 3:15 Xin RongWord Embedding Explained and Visualized 3:30 – 4:15 Daniel PresselDeep Networks for Common NLP Tasks 4:30 – 5:15 Catherine Finegan-Dollak Introduction to LSTMs 5:30 – 6:15 Gint Puskorius Opportunities in Automotive Applications of Deep Learning at Ford Motor Company 6:30 – 7:15 Honglak Lee – Keynote

Details

Date:
November 7, 2015
Time:
1:30 pm - 7:30 pm