AI Week 2018 KR Meets ML Workshop       

links

AI Week 2018 KR Meets ML Workshop - overview


KR Meets ML Workshop

IBM AI Research Week 2018

Logistics

Date: Thursday, Oct 4th 2018
Location: IBM Research, 75 Binney St, Cambridge MA (Room: 2B34)
Location: MIT Samberg Conference Center, Chang Building (E52)
Room: Dining Room 4 (6th Floor)
Remote: ibm.biz/krml-webex (opens at 9am)
Time: 09:00 - 16:00 

Note (Location Change): The workshop has moved to the MIT Samberg Conference Center, Chang Building (E52). The workshop will be held in DR4 (Dining Room 4) on the Sixth Floor. Please use the links to navigate to the right location.

Description

The incredible progress in machine learning over the past few years has led to solving problems that have challenged AI researchers for decades. The surge of enthusiasm for these ML breakthroughs have caused some to declare (yet again) that it is the end of “knowledge representation” as AI moves into a world dominated by neural networks, data mining and the knowledge graph. This workshop is focused on exploring the intersection/union of these breakthrough methods and the outstanding problems that require a level of knowledge and reasoning beyond what is currently available in these learning and representation techniques. Topics of interest include: problems that require integration of learning and symbolic knowledge; how ML might take advantage of complex, interconnected background knowledge; how knowledge representation systems might be built that can have as input the results of modern learning technologies; and how those problems that remain unsolved might be tackled.

Program / Schedule

Keynote

Leslie Valiant, Harvard

Schedule (Tentative)

09:00 - 09:10
Workshop Welcome: Kartik Talamadupula
09:10 - 09:30
Invited Talk: Background Knowledge
Jim Hendler, Rensselaer Polytechnic Institute
09:30 - 10:30
KeynoteWhat Needs to be Added to Machine Learning?
Leslie Valiant, Harvard University
10:30 - 11:00
Coffee Break
11:00 - 11:20
Invited TalkReasoning and Common Sense with Embedded Knowledge Graphs
Andrew McCallum, Univ of Massachusetts Amherst
11:20 - 11:40
Invited TalkProspects for Unifying Learning and Careful Reasoning
Michael J. Witbrock, IBM Research AI
11:40 - 12:00
Invited TalkKnowledge Representation and Machine Learning... A Marriage for Modern Times?
Deborah L. McGuinness, Rensselaer Polytechnic Institute
12:00 - 12:30
Lightning Talks (5 Minutes Each)
  • Naveen Sundar Govindarajulu (RPI): On Evaluating Learning to Reason
  • Hendrik Strobelt (IBM Research): The Role of Visualization in ML & KR
  • Sarah Keren (Harvard): Goal Recognition Design
  • Satyajeet Raje (IBM Global CDO): Enabling Intelligent Enterprise Data Lakes: Role of Knowledge Graphs and ML
  • Matt Davis (Invitae): Reasoning with Task-Specific Graph Embeddings of Biological Networks
  • Nicholas Mattei (IBM Research AI): Complex QA as a Unifying Task: From Knowledge Graphs to Attention Models
12:30 - 14:00
Lunch Break
14:00 - 15:30
Break-out Sessions / BoF (see detailed agenda)
  • What can ML <-> KR do for each other?
  • Evaluation of Progress in the Area
  • Hard Long-Term Problems
15:30 - 16:00
Workshop Wrap-up


Please visit the detailed agenda for more details on the schedule.

 

Organizers

  • Kartik Talamadupula, IBM Research AI
  • Jim Hendler, RPI 
  • Michael Witbrock, IBM Research AI

Local Organization Chairs

  • Nicholas Mattei, IBM Research AI (Yorktown Heights)
  • Christian Muise, IBM Research AI (Cambridge)
  • Veronika Thost, IBM Research AI (Cambridge)

Registration

Registration to the workshop is now closed.