WebML WG Teleconference – 8 September 2022 - 14:00-15:00 UTC

Latest version: https://github.com/webmachinelearning/meetings/blob/main/telcons/2022-09-08-wg-agenda.md


WebML WG Teleconference – 8 September 2022 - 14:00-15:00 UTC
San Francisco (U.S.A. - California)     Thu, 8 September 2022   07:00   UTC-7 hours
Boston (U.S.A. - Massachusetts) Thu, 8 September 2022   10:00   UTC-4 hours
London (United Kingdom - England)       Thu, 8 September 2022   15:00   UTC+1 hours
Berlin (Germany)        Thu, 8 September 2022   16:00   UTC+2 hours
Helsinki (Finland)      Thu, 8 September 2022   17:00   UTC+3 hours
Shanghai (China)        Thu, 8 September 2022   22:00   UTC+8 hours
Tokyo (Japan)   Thu, 8 September 2022   23:00   UTC+9 hours
Corresponding UTC (GMT) Thu, 8 September 2022   14:00 UTC

Other locations: https://www.timeanddate.com/worldclock/fixedtime.html?iso=20220908T14


<https://github.com/webmachinelearning/meetings/blob/main/telcons/2022-09-08-wg-agenda.md#logistics>Logistics

  *   Chair: Anssi
  *   Scribe: ? (see howto<https://github.com/webmachinelearning/meetings/blob/main/scribe-howto.md>, volunteers welcome!)
  *   IRC: irc://irc.w3.org:6667/#webmachinelearning

  *   IRC web client: https://irc.w3.org/?channels=#webmachinelearning

  *   Joining instructions: https://lists.w3.org/Archives/Member/internal-webmachinelearning/2020Apr/0000.html

  *   Minutes: https://www.w3.org/2022/09/08-webmachinelearning-minutes.html


<https://github.com/webmachinelearning/meetings/blob/main/telcons/2022-09-08-wg-agenda.md#%E2%84%B9%EF%B8%8F-webnn-api-candidate-recommendation-open-issues>ℹ️ WebNN API Candidate Recommendation open issues
Review and discuss the current CR issues, work out a plan to address the issues prior to the expected CR publication.

  *   Current CR issues https://github.com/webmachinelearning/webnn/issues?q=is%3Aissue+is%3Aopen+label%3Acr


  *   Support asynchronous context creation (two PRs for two design alternatives)

     *    issue: webmachinelearning/webnn#272<https://github.com/webmachinelearning/webnn/issues/272>
     *    PR: webmachinelearning/webnn#274<https://github.com/webmachinelearning/webnn/pull/274>
     *    PR: webmachinelearning/webnn#285<https://github.com/webmachinelearning/webnn/pull/285>
  *   Define ULP (unit of least precision) tolerances for testing

     *    issue: webmachinelearning/webnn#265<https://github.com/webmachinelearning/webnn/issues/265>
     *    issue (exp op DML/baseline data): webmachinelearning/webnn#288<https://github.com/webmachinelearning/webnn/issues/288>
     *    PR: web-platform-tests/wpt#34287<https://github.com/web-platform-tests/wpt/pull/34287>
  *   Add method steps and normative algorithms to operations

     *    issue: webmachinelearning/webnn#210<https://github.com/webmachinelearning/webnn/issues/210>
     *    issue: webmachinelearning/webnn#211<https://github.com/webmachinelearning/webnn/issues/211>
  *   Support for int8 quantized models

     *    issue: webmachinelearning/webnn#128<https://github.com/webmachinelearning/webnn/issues/128>

<https://github.com/webmachinelearning/meetings/blob/main/telcons/2022-09-08-wg-agenda.md#%E2%84%B9%EF%B8%8F-use-cases-for-future-work>ℹ️ Use cases for future work

  *   Content filtering

     *    PR: webmachinelearning/webnn#253<https://github.com/webmachinelearning/webnn/pull/253>
     *    future work proposal: webmachinelearning/proposals#4<https://github.com/webmachinelearning/proposals/issues/4>
  *   Performance adaptation

     *    PR: webmachinelearning/webnn#207<https://github.com/webmachinelearning/webnn/pull/207>
     *    future work proposal or close?

<https://github.com/webmachinelearning/meetings/blob/main/telcons/2022-09-08-wg-agenda.md#%E2%84%B9%EF%B8%8F-webgpu-working-group-review-request-and-webnn-webgpu-interop>ℹ️ WebGPU Working Group review request and WebNN-WebGPU interop
Discuss WG's response to the WebGPU review request, focus on WebNN-WebGPU interop requirements and issues.

  *   Review request https://lists.w3.org/Archives/Public/public-webmachinelearning-wg/2022Jul/0000.html


  *   11 August 2022 discussion https://www.w3.org/2022/08/11-webmachinelearning-minutes.html#t08


  *    related issue: webmachinelearning/webnn#264<https://github.com/webmachinelearning/webnn/issues/264>

<https://github.com/webmachinelearning/meetings/blob/main/telcons/2022-09-08-wg-agenda.md#%E2%84%B9%EF%B8%8F-proposed-new-features>ℹ️ Proposed new features

  *    issue: Need for type casting? webmachinelearning/webnn#284<https://github.com/webmachinelearning/webnn/issues/284>

  *    issue: Softmax should only support input of floating-point types webmachinelearning/webnn#283<https://github.com/webmachinelearning/webnn/issues/283>

Revisit proposed new features discussed earlier:

  *    proposed CR issue: Should MLBufferView + MLOperandDescriptor be strongly typed webmachinelearning/webnn#275<https://github.com/webmachinelearning/webnn/issues/275>

  *    proposed CR issue: Support coordinate transformation modes for Resample2d webmachinelearning/webnn#270<https://github.com/webmachinelearning/webnn/issues/270>

  *    issue: Define the data type of the padding, strides and dilations webmachinelearning/webnn#269<https://github.com/webmachinelearning/webnn/issues/269>

  *    issue: l2Pool2d algorithm clarifications webmachinelearning/webnn#278<https://github.com/webmachinelearning/webnn/issues/278>

<https://github.com/webmachinelearning/meetings/blob/main/telcons/2022-09-08-wg-agenda.md#%E2%84%B9%EF%B8%8F-webml-wg-charter-2023-2025-early-heads-up>ℹ️ WebML WG Charter 2023-2025 early heads-up
WebML WG Charter 2023-2025 brainstorming discussion. Start solicit use cases and model requirements for WebNN "V2". WG rechartering process to formally kick off early 2023. New technical proposals to be incubated in the CG prior to the WG adoption.

  *   WG Charter: https://www.w3.org/2021/04/web-machine-learning-charter.html

  *   CG Charter: https://webmachinelearning.github.io/charter/