- From: Kostiainen, Anssi <anssi.kostiainen@intel.com>
- Date: Thu, 8 Apr 2021 09:10:48 +0000
- To: "public-webmachinelearning@w3.org" <public-webmachinelearning@w3.org>
- Message-ID: <E6394643-6F37-480A-8FAE-B0A076ABD195@intel.com>
Latest version: https://github.com/webmachinelearning/meetings/blob/master/telcons/2021-04-15-agenda.md
WebML CG Teleconference – 15 April 2021 - 14:00-15:00 UTC+0
San Francisco (U.S.A. - California) Thu, 15 April 2021 07:00 UTC-7 hours
Boston (U.S.A. - Massachusetts) Thu, 15 April 2021 10:00 UTC-4 hours
London (United Kingdom - England) Thu, 15 April 2021 15:00 UTC+1 hours
Berlin (Germany) Thu, 15 April 2021 16:00 UTC+2 hours
Helsinki (Finland) Thu, 15 April 2021 17:00 UTC+3 hours
Shanghai (China) Thu, 15 April 2021 22:00 UTC+8 hours
Tokyo (Japan) Thu, 15 April 2021 23:00 UTC+9 hours
Corresponding UTC (GMT) Thu, 15 April 2021 14:00 UTC
Other locations: https://www.timeanddate.com/worldclock/fixedtime.html?iso=20210415T14
<https://github.com/webmachinelearning/meetings/blob/master/telcons/2021-04-15-agenda.md#logistics>Logistics
* Chair: Anssi
* Scribe: ?
* IRC: irc://irc.w3.org:6667/#webmachinelearning
* IRC web client: http://irc.w3.org/?channels=#webmachinelearning
* Call-in details: https://lists.w3.org/Archives/Member/internal-webmachinelearning/2020Apr/0000.html
* Minutes: https://www.w3.org/2021/04/15-webmachinelearning-minutes.html
<https://github.com/webmachinelearning/meetings/blob/master/telcons/2021-04-15-agenda.md#agenda>Agenda
<https://github.com/webmachinelearning/meetings/blob/master/telcons/2021-04-15-agenda.md#web-machine-learning-wg>Web Machine Learning WG
Discuss changes to the proposed Web Machine Learning Working Group charter:
https://lists.w3.org/Archives/Public/public-webmachinelearning/2021Apr/0002.html
Please contact Dom if any of these changes feel problematic, before April 16.
<https://github.com/webmachinelearning/meetings/blob/master/telcons/2021-04-15-agenda.md#privacy-review-feedback-normative-changes>Privacy review feedback normative changes
[privacy-tracker] Make WebNN API a policy-controlled feature: https://github.com/webmachinelearning/webnn/issues/145
Review PR: https://github.com/webmachinelearning/webnn/pull/159
<https://github.com/webmachinelearning/meetings/blob/master/telcons/2021-04-15-agenda.md#operation-specific-apis-proposal>Operation-specific APIs proposal
Background: operation-specific APIs proposal (Ping, Jonathan): https://github.com/webmachinelearning/proposals/issues/2
Continue refine work-in-progress features that satisfy the requirements of this proposal:
* WebAssembly scenario of the op level execution use case (Ningxin): https://github.com/webmachinelearning/webnn/issues/156
* Clarify at which point weights are used in the compilation (Rafael): https://github.com/webmachinelearning/webnn/issues/157
* Review the PR to address #157: WIP (Ningxin)
* Discuss how the caller using an op-specific API can resource upload and download (Chai)
<https://github.com/webmachinelearning/meetings/blob/master/telcons/2021-04-15-agenda.md#tag-review-feedback---open-issues-without-associated-prs>TAG review feedback - open issues without associated PRs
Review proposed resolutions and provide your comments:
* Define a common term for logical tensor changes?
* Issue: https://github.com/webmachinelearning/webnn/issues/150
* Proposed resolution: TBD
* Isomorphic JS story, worker scope exposure?
* Issue: https://github.com/webmachinelearning/webnn/issues/142
* Proposed resolution: Expose ML interface to Worker in addition to its default Window context
* Ergonomics of the JS examples
* Issue: https://github.com/webmachinelearning/webnn/issues/139
* Proposed resolution: Close with a note the primary API consumer is a JS framework
* String enum for activations
* Issue: https://github.com/webmachinelearning/webnn/issues/138
* Proposed resolution: Close with a note the current organization of activation functions improves developer ergonomics
Received on Thursday, 8 April 2021 09:11:12 UTC