WebML WG Teleconference – 8 May 2025 - 15:00-16:00 UTC

Latest version: https://github.com/webmachinelearning/meetings/blob/main/telcons/2025-05-08-wg-agenda.md


WebML WG Teleconference – 8 May 2025 - 15:00-16:00 UTC
<https://github.com/webmachinelearning/meetings/blob/main/telcons/2025-05-08-wg-agenda.md#webml-wg-teleconference--8-may-2025---1500-1600-utc>
See the timezone table ...
San Francisco   Thu, 8 May 2025 08:00
Boston  Thu, 8 May 2025 11:00
London  Thu, 8 May 2025 16:00
Berlin  Thu, 8 May 2025 17:00
Helsinki        Thu, 8 May 2025 18:00
Shanghai        Thu, 8 May 2025 23:00
Tokyo   Fri, 9 May 2025 00:00
UTC     Thu, 8 May 2025 15:00 UTC

Other locations: https://www.timeanddate.com/worldclock/fixedtime.html?iso=20250508T15


Logistics
<https://github.com/webmachinelearning/meetings/blob/main/telcons/2025-05-08-wg-agenda.md#logistics>

  *   Chair: Anssi
  *   Scribe: Anssi
  *   IRC: irc://irc.w3.org:6667/#webmachinelearning

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

  *   Zoom joining instructions: https://lists.w3.org/Archives/Member/internal-webmachinelearning/2023Jun/0000.html

  *   Minutes: https://www.w3.org/2025/05/08-webmachinelearning-minutes.html


Agenda
<https://github.com/webmachinelearning/meetings/blob/main/telcons/2025-05-08-wg-agenda.md#agenda>
🧪 Incubations
<https://github.com/webmachinelearning/meetings/blob/main/telcons/2025-05-08-wg-agenda.md#-incubations>

Discuss recent WebML Community Group development to keep the Working Group abreast of incubation progress.

New proposal: ✨ Local Inference Web extension webmachinelearning/proposals#9<https://github.com/webmachinelearning/proposals/issues/9>

🏷️ Operator specific issues
<https://github.com/webmachinelearning/meetings/blob/main/telcons/2025-05-08-wg-agenda.md#%EF%B8%8F-operator-specific-issues>

Review and discuss operator specific issues that reduce code complexity and improve maintainability, e.g.:

  *   layerNormalization

     *   ⨀ webmachinelearning/webnn#748<https://github.com/webmachinelearning/webnn/issues/748>
  *   triangular

     *   ⨀ webmachinelearning/webnn#768<https://github.com/webmachinelearning/webnn/issues/768>
  *   sign

     *   ⨀ webmachinelearning/webnn#845<https://github.com/webmachinelearning/webnn/issues/845>

ℹ️ WebNN wide review
<https://github.com/webmachinelearning/meetings/blob/main/telcons/2025-05-08-wg-agenda.md#%E2%84%B9%EF%B8%8F-webnn-wide-review>

Review and discuss interim wide review feedback and the group's proposed response.

  *   ☰ webmachinelearning/webnn#239 (comment)<https://github.com/webmachinelearning/webnn/issues/239#issuecomment-2740740891>

ℹ️ Explainer updates: WebNN, MLTensor, MLGraph caching
<https://github.com/webmachinelearning/meetings/blob/main/telcons/2025-05-08-wg-agenda.md#%E2%84%B9%EF%B8%8F-explainer-updates-webnn-mltensor-mlgraph-caching>

Discuss recent explainer updates and suggested improvements.

  *   ☰ WebNN explainer https://github.com/webmachinelearning/webnn/blob/main/explainer.md

     *   ⨀ webmachinelearning/webnn#840<https://github.com/webmachinelearning/webnn/issues/840>
  *   ☰ MLTensor explainer https://github.com/webmachinelearning/webnn/blob/main/mltensor-explainer.md

     *   ⛙ webmachinelearning/webnn#844<https://github.com/webmachinelearning/webnn/pull/844>
  *   ☰ Caching mechanism for MLGraph explainer

ℹ️ Query supported devices
<https://github.com/webmachinelearning/meetings/blob/main/telcons/2025-05-08-wg-agenda.md#%E2%84%B9%EF%B8%8F-query-supported-devices>

Discuss the query supported devices feature, now split in two:

  *   ⨀ Before graph compilation webmachinelearning/webnn#815<https://github.com/webmachinelearning/webnn/issues/815>
  *   ⨀ After graph compilation webmachinelearning/webnn#836<https://github.com/webmachinelearning/webnn/issues/836>

ℹ️ Core operator set
<https://github.com/webmachinelearning/meetings/blob/main/telcons/2025-05-08-wg-agenda.md#%E2%84%B9%EF%B8%8F-core-operator-set>

Revisit our core operator set effort that aims to identify current primitive gaps by mapping compositional fundamentals to WebNN operators.

Discuss any new information on rounding behavior across backends to understand feasibility for inclusion into the core operator set to help with e.g. quantization decomposition.

  *   ⨀ webmachinelearning/webnn#573<https://github.com/webmachinelearning/webnn/issues/573>
  *   ☰ Machine Learning Operator Mapping - All Raw Operators<https://onedrive.live.com/edit?id=EE82F5C6F06C7371!345450&resid=EE82F5C6F06C7371!345450&ithint=file%2Cxlsx&authkey=!AK8f-RDTleqlLXE&wdo=2&cid=ee82f5c6f06c7371>