imported>KennethBaclawski at 02:53, 11 October 2018

2018-10-11T02:53:11Z

New page

{| class="wikitable" style="float:right; margin-left: 10px;" border="1" cellpadding="10"
|-
! scope="row" | Session
| [[session::Planning]]
|-
! scope="row" | Duration
| [[duration::1 hour]]
|-
! scope="row" rowspan="3" | Date/Time
| [[has date::October 10 2018 16:00 GMT]]
|-
| 9:00am PDT/12:00pm EDT
|-
| 5:00pm BST/6:00pm CST
|-
! scope="row" | Convener
| [[convener::KenBaclawski]]
|}

= [[OntologySummit2019|Ontology Summit 2019]] Planning =
{{:OntologySummit2019/Theme}}

== Agenda ==
* Discuss and improve the theme description
* Select tracks and co-champions

== Conference Call Information ==
* Date: '''Wednesday, 10-October-2018'''
* Start Time: 9:00am PDT / 12:00pm EDT / 6:00pm CEST / 5:00pm BST / 1600 UTC
** ref: [http://www.timeanddate.com/worldclock/fixedtime.html?month=10&day=10&year=2018&hour=12&min=00&sec=0&p1=179 World Clock]
* Expected Call Duration: 1 hour
* The Video Conference URL is https://zoom.us/j/689971575
** iPhone one-tap :
*** US: +16699006833,,689971575# or +16465588665,,689971575#
** Telephone:
*** Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 558 8665
*** Meeting ID: 689 971 575
*** International numbers available: https://zoom.us/u/Iuuiouo
* [http://bit.ly/2MRyySS Chat Room]

== Attendees ==
* [[AlexShkotin|Alex Shkotin]]
* [[JanetSinger|Janet Singer]]
* [[JohnSowa|John Sowa]]
* [[KenBaclawski|Ken Baclawski]]
* [[MarkUnderwood|Mark Underwood]]
* [[RamSriram|Ram D. Sriram]]
* [[ToddSchneider|Todd Schneider]]

== Proceedings ==
[12:08] KenBaclawski: Zachary C. Lipton, The Mythos of Model Interpretability, Comm. ACM 61(10)36-43. 2018. DOI: 10.1145/3233231

[12:08] janet singer: https://arxiv.org/abs/1606.03490

[12:12] KenBaclawski: New title for the summit: Explainability.

[12:12] KenBaclawski: Ram suggests "Explainability of Reasoning"

[12:15] KenBaclawski: Meeting page is http://ontologforum.org/index.php/ConferenceCall_2018_10_10

[12:15] Ram D. Sriram: My suggestion The Role of Ontologies in Explaining Reasoning

[12:21] ToddSchneider: Having a short 'catchy' (and vague) title is an easier sell.

[12:21] KenBaclawski: John suggests that explanations answer the question "Why?". A follow up question is "What do I do next?"

[12:21] AlexShkotin: @Ram +1

[12:24] ToddSchneider: Ken, I have a few tweaks for the [explanation] text on the planning page. Shall I email them to you?

[12:24] KenBaclawski: John: Explanation is a process or dialog. It can be between people and systems or between two systems, i.e., agent to agent.

[12:25] KenBaclawski: @Todd: Yes, please email your tweaks.

[12:28] John Sowa: General principle: An explanation is an answer to a follow-up question.

[12:32] AlexShkotin: In axiomatic theory there are 3 kind of explanation: axiom, definition and proof

[12:32] John Sowa: Example: Current telephone answering systems don't give explanations. But an XAI system would behave like a human operator who could answer questions.

[12:38] janet singer: Maybe start with "An explanation is always an answer to a question"

[12:49] AlexShkotin: @Ram, may be "The Role of Ontologies in Explaining" is better

[13:00] Ram D. Sriram: Prof. Louiqa Rashid fromm Univ of Maryland held a track on Finance at the OKN workshop. She may a person to recruit for the finance domain.

[13:00] Mark Underwood: I've been working to summarize work in this paper, which isn't AI per se, but has evidence-based analysis

Explanation in artificial intelligence: Insights from the social sciences, Aug. 2018. [Online]. Available: http://arxiv.org/abs/1706.07269

[13:01] Mark Underwood: The use case family I have some access to is regulation about Fair Credit reporting, which requires that credit decisions result in explanations to consumers

[13:02] Mark Underwood: Often those decisions of late are ML-generated, and in near-real time

[13:02] Mark Underwood: If a track, I can help frame it and try to invite others to speak

[13:02] Mark Underwood: "Explanation in artificial intelligence: Insights from the social sciences," Aug. 2018. [Online]. Available: http://arxiv.org/abs/1706.07269

[13:03] Mark Underwood: The latter paper is ontology-poor but touches on some relevant XAI issues that can be rolled into use cases. Would require some creative thinking

[13:03] ToddSchneider: Have to go. Thank you.

[13:05] Mark Underwood: In some IEEE groups we are also wrestling with explanations for children, vision impaired, autistic, etc.

[13:05] Mark Underwood: There's an assumption about the likely audience that is often oversimplified -- or just wrong

Ken Baclawski: I will add the suggestions to the summit description, and I will post a call for summit co-champions.

Ken Baclawski: Two tracks have been proposed so far: Mark Underwood on Financial Explanations, Ram D. Sriram on Medical Explanations.

== Resources ==
[http://bit.ly/2RGNTIg Audio Recording]

== Previous Meetings ==
{{#ask: [[Category:OntologySummit2019]] [[Category:Icom_conf_Conference]] [[has date::<<{{#show:{{PAGENAME}}|?has date}}]] |?|
mainlabel=-|sort=has date |order=desc| format=ul| limit=3}}

== Next Meetings ==
{{#ask: [[Category:OntologySummit2019]] [[Category:Icom_conf_Conference]] [[has date::>>{{#show:{{PAGENAME}}|?has date}}]] |?|
mainlabel=-|sort=has date |order=asc| format=ul| limit=3}}

[[Category:OntologySummit2019]]
[[Category:Icom_conf_Conference]]
[[Category:Occurrence| ]]

ConferenceCall 2018 10 10 - Revision history

imported>KennethBaclawski at 02:53, 11 October 2018