Recording of this event is available at https://ibm.biz/nsworkshop2023-playback

A new era of AI is rapidly emerging: neuro-symbolic AI combines knowledge-driven, symbolic AI with data-driven machine learning approaches. IBM is a leader in the research and development of neuro-symbolic AI technologies and we invite graduate students, AI practitioners, and anyone interested in this emerging field to participate in the IBM Neuro-Symbolic AI Workshop 2023, that will be held on January 23-27 as a virtual event.

This event is a follow-on to the 1st workshop on the topic we held in January 2022. We also organized a summer school in August of 2022. These events showcased the breadth and depth of the work being done in this field at IBM and by our collaborators. Participation in the previous events is not a prerequisite for attending the upcoming workshop. All the talks in the workshop are meant to be self-contained. To access the previous events agenda page and recordings, visit:

IBM Neuro-Symbolic AI Workshop 2022 (http://ibm.biz/ns-wkshp)
IBM Neuro-Symbolic AI Summer School 2022 (http://ibm.biz/nsss2022)
To get familiar with IBM’s neuro-symbolic approach, you can get a free IBM Neuro-Symbolic AI Essentials digital badge (https://www.credly.com/org/ibm/badge/neuro-symbolic-ai-essentials)

What is Neuro-Symbolic AI? Compared to pure neural approaches, our key goals for a neuro-symbolic AI system include:

Learning with less and zero-shot learning; we hypothesize this can be aided by symbolic human knowledge.
Generalization of the solutions to unseen tasks and unforeseen data distributions; we hypothesize this can be aided by reasoning.
Explainability by construction; we hypothesize this can be aided by human readable knowledge and step-by-step explanation of the reasoning process.

The event is organized around 5 key questions of neuro-symbolic AI. We will address 1 question per day. The workshop will include over 15 IBM talks, and 5 panels in various areas of theory and the application of neuro-symbolic AI. We will also have over 15 distinguished external speakers to share an overview of neuro-symbolic AI and its history. This is a virtual only event and the registration for the event is free. The registered participants will get access to the recording of all sessions after the event.

Session and Time	Topic
Opening 08:45 - 9:00	Goals Open problems Event overview
Topic Introduction 9:00 - 9:10	TBD
Invited talk 9:10 - 9:35
IBM Talk 9:35 - 10:00	An expressive probabilistic logic that generalizes prior formalisms that combine logic and probability. Given imprecise information represented by probability bounds and conditional probability bounds on logic formulas, an LCN specifies a set of probability distributions over all its interpretations. Our approach allows propositional and first-order logic formulas with few restrictions, e.g., without requiring acyclicity. We also define a generalized Markov condition that allows us to identify implicit independence relations between atomic formulas. We evaluate our method on benchmark problems such as random networks, Mastermind games with uncertainty and credit card fraud detection. Our results show that the LCN outperforms existing approaches; its advantage lies in aggregating multiple sources of imprecise information.
Break 10:00 - 10:10
IBM Talk 10:10 - 10:35	RV theory
Invited talk 10:35 - 11:00	TBD
Break 11:00 - 11:10
Invited talk 11:10 - 11:35	TBD
IBM Talk 11:35 - 12:00	TBD
Break 12:00 - 12:10
Panel 12:10 - 13:00	TBD
Pre-recorded short talks	Recording: Play Recording Recording: Play Recording Recording: Play Recording Recording: Play Recording Recording: Play Recording Recording: Play Recording Recording: Play Recording Recording: Play Recording

Session and Time	Topic
Opening 9:00 - 9:02	Introduction to short talks
Topic Introduction 9:02 - 9:10	TBD
IBM talk 9:10 - 9:35	What are the characteristics of useful neuro-symbolic benchmarks? Interactive fiction as a challenging benchmark for Neuro-symbolic AI
IBM talk 9:35 - 10:00	TBD
Break 10:00 - 10:10
IBM talk 10:10 - 10:35	TBD
Invited talk 10:35 - 11:00	will cover research on Story Understanding too
Break 11:00 - 11:10
Invited talk 11:10 - 11:35	TBD
Invited talk 11:35 - 12:00	TBD
Break 12:00 - 12:10
Panel 12:10 - 13:00	TBD
Pre-recorded short talks	Recording: Play Recording Recording: Play Recording Recording: Play Recording

Session and Time	Topic
Opening 9:00 - 9:02	LLM open problems Introduction to short talks
Topic Introduction 9:02 - 9:10	TBD
Invited talk 9:10 - 9:35	TBD
Invited talk 9:35 - 10:00	TBD
Break 10:00 - 10:10
IBM talk 10:10 - 10:35	TBD
IBM talk 10:35 - 11:00	TBD
Break 11:00 - 11:10
Invited talk 11:10 - 11:35	TBD
IBM talk 11:35 - 12:00	TBD
Break 12:00 - 12:10
Panel 12:10 - 13:00	TBD
Pre-recorded short talks	Recording: Play Recording Recording: Play Recording Recording: Play Recording

Session and Time	Topic
Opening 9:00 - 9:02	Introduction to short talks
Topic Introduction 9:02 - 9:10	TBD
Invited talk 9:10 - 9:35	One of the central problems in AI has been the knowledge acquisition bottleneck. Progress has been made in multiple areas, as the existence of billion-fact knowledge graphs illustrates, but many open challenges remain. This talk will discuss some of these challenges, the two different categories of approaches that are being used to tackle them, and how we need to rethink knowledge acquisition as we start fielding AI systems more broadly.
IBM talk 9:35 - 10:00	ULKB: beyond graph composition. To support NLP applications we need the federation of linguistic, ontologic and logic knowledge Parsing of NL with MRS and AMR and its knowledge requirements ULKB logic and rules to help with contextual linking
Break 10:00 - 10:10
Invited talk 10:10 - 10:35	TBD
Invited talk 10:35 - 11:00	TBD
Break 11:00 - 11:10
IBM talk 11:10 - 11:35	TBD
Invited talk 11:35 - 12:00	TBD
Break 12:00 - 12:10
Panel 12:10 - 13:00	TBD
Pre-recorded short talks	Recording: Play Recording ... Recording: Play Recording

IBM Neuro-Symbolic AI Workshop January 23-27, 2023

Agenda

Invited Speakers

IBM Speakers

Organizers

Webpage Creators

IBM Neuro-Symbolic AI Workshop
January 23-27, 2023