30 October 2018Edinburgh & Lothians
Venue: Auditorium, University of Edinburgh Business School, 29 Buccleuch Place, Edinburgh
Time: 18:30 to 20:00
Advances in artificial intelligence bring about many new, interesting application now tackled with often deep learning architectures. From customer profiling, over speech recognition to image classification. These tasks, however, are still focusing on a single problem, however, deep learning can also be applied to multi-purpose applications as well.
In this seminar, Bryan McCann, Senior Research Scientist at Salesforce, presents the Natural Language Decathlon (decaNLP), a challenge that spans ten tasks: question answering, machine translation, summarization, natural language inference, sentiment analysis, semantic role labeling, zero-shot relation extraction, goal-oriented dialogue, semantic parsing, and common sense pronoun resolution by making use of a Multitask Question Answering Network (MQAN).
This talk will be followed by a Q&A with Bryan.
Coffee and registration from 18:00. The event will be followed by a networking drinks reception.
About the speaker
Bryan McCann is a Research Scientist at Salesforce. He focuses on transfer learning and multitask learning for natural language processing. Most recently, Bryan proposed the Natural Language Decathlon (decaNLP) and a Multitask Question Answering Network to tackle all ten tasks in decaNLP. Before decaNLP, he showed that the intermediate representations, or context vectors (CoVe), of machine translation systems carry information that aids learning in question answering and text classification systems. Prior to working at Salesforce, Bryan studied at Stanford University, where he completed a BS and MS in Computer Science as well as a BA in Philosophy.
Salesforce Research advances state-of-the-art AI techniques, developing models, prototypes, and experiments that pave the path for innovative products on the Einstein AI Platform. We tackle real-world problems from Salesforce's enterprise customers harnessing the latest deep learning models, from image recognition to natural language processing.