WebML CG Teleconference – 3 September 2020 - 14:00-15:00 UTC+0

https://github.com/webmachinelearning/meetings/blob/master/telcons/2020-09-03-agenda.md


WebML CG Teleconference – 3 September 2020 - 14:00-15:00 UTC+0
San Francisco (U.S.A. - California)     Thu, 3 September 2020   07:00 PDT       UTC-7 hours
Boston (U.S.A. - Massachusetts) Thu, 3 September 2020   10:00 EDT       UTC-4 hours
London (United Kingdom - England)       Thu, 3 September 2020   15:00 BST       UTC+1 hours
Berlin (Germany)        Thu, 3 September 2020   16:00 CEST      UTC+2 hours
Helsinki (Finland)      Thu, 3 September 2020   17:00 EEST      UTC+3 hours
Shanghai (China)        Thu, 3 September 2020   22:00 CST       UTC+8 hours
Tokyo (Japan)   Thu, 3 September 2020   23:00 JST       UTC+9 hours
Corresponding UTC (GMT) Thu, 3 September 2020   14:00 UTC

Other locations: https://www.timeanddate.com/worldclock/fixedtime.html?iso=20200903T14


<https://github.com/webmachinelearning/meetings/blob/master/telcons/2020-09-03-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 (login with your W3C account<https://www.w3.org/Help/Account/>) - Zoom app install<https://zoom.us/download> recommended
  *   Minutes: https://www.w3.org/2020/09/03-webmachinelearning-minutes.html


<https://github.com/webmachinelearning/meetings/blob/master/telcons/2020-09-03-agenda.md#agenda>Agenda
<https://github.com/webmachinelearning/meetings/blob/master/telcons/2020-09-03-agenda.md#workshop-presentations-and-gh-discussions>Workshop presentations and GH discussions

Workshop presentations are actively being discussed on GitHub. CG participants feedback has been appreciated, further engagement welcome.

https://www.w3.org/2020/06/machine-learning-workshop/presentations.html


https://github.com/w3c/machine-learning-workshop/issues


<https://github.com/webmachinelearning/meetings/blob/master/telcons/2020-09-03-agenda.md#webnn-polyfill-and-samples>WebNN polyfill and samples

Discuss review feedback and suggestions for the foundational implementation

https://github.com/webmachinelearning/webnn-polyfill/pull/1


<https://github.com/webmachinelearning/meetings/blob/master/telcons/2020-09-03-agenda.md#gru-and-corresponding-ops>GRU and corresponding ops

Fill the operator gaps to support noise suppression first-wave models

Adding GRU and GRUcell operators to support GRU recurrent network. Defining a cell operator in addition to the network operator for added customization flexibility e.g. to support stacked cell recurrent network, etc.

Review PR: https://github.com/webmachinelearning/webnn/pull/83


Proposal to close noise suppression issue with PR #83:

https://github.com/webmachinelearning/webnn/issues/66


<https://github.com/webmachinelearning/meetings/blob/master/telcons/2020-09-03-agenda.md#first-wave-models-and-ops-delta-with-webnn-api-definition>First-wave models and ops delta with WebNN API definition

Review the delta between the first-wave models<https://github.com/webmachinelearning/webnn/blob/master/op_compatibility/first_wave_models.md> ops and WebNN API<https://webmachinelearning.github.io/webnn/#api> surface:

  *    clamp
  *    globalAveragePool - Lowering to reducewindow, add and div?
  *    gru
  *    sigmoid
  *    split - Lowering to slice?
  *    squeeze

Discuss which ops to add to the spec definition considering e.g. major platform support and performance implications.