WebML WG Teleconference – 7 September 2023 - 14:00-15:00 UTC

Latest version: https://github.com/webmachinelearning/meetings/blob/main/telcons/2023-09-07-wg-agenda.md


WebML WG Teleconference – 7 September 2023 - 14:00-15:00 UTC
See the timezone table ...
<https://github.com/webmachinelearning/meetings/blob/main/telcons/2023-09-07-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/2023Jun/0000.html

  *   Minutes: https://www.w3.org/2023/09/07-webmachinelearning-minutes.html


<https://github.com/webmachinelearning/meetings/blob/main/telcons/2023-09-07-wg-agenda.md#agenda>Agenda
<https://github.com/webmachinelearning/meetings/blob/main/telcons/2023-09-07-wg-agenda.md#%E2%84%B9%EF%B8%8F-webnn-v2-proposed-text-to-text-and-text-generation-model-targets>ℹ️ WebNN v2: Proposed text-to-text and text-generation model targets

Review, discuss and choose appropriate text-to-text generation (encoder-decoder) and text-generation (decoder-only) v2 model targets informed by Transformers.js and Hugging Face Hub usage trends.

Review and discuss feedback for the proposed new ops and data types informed by WebNN DirectML backend prototyping efforts.

Details:

  *   Text-to-text and text-gen v2 model candidates: webmachinelearning/webnn#375 (comment)<https://github.com/webmachinelearning/webnn/issues/375#issuecomment-1693955191>
  *   Proposed new v2 ops and data types: webmachinelearning/webnn#375 (comment)<https://github.com/webmachinelearning/webnn/issues/375#issuecomment-1674224992>

Background material:

  *   Transformers.js presentation by Joshua Lochner<https://lists.w3.org/Archives/Public/www-archive/2023Jun/att-0000/Transformers_js.pdf>
  *   Transformer models presentation by Dwayne Robinson<https://lists.w3.org/Archives/Public/www-archive/2023Jun/att-0005/2023-06-29_WebNN_and_Transformers_Progress_W3C.pdf>

<https://github.com/webmachinelearning/meetings/blob/main/telcons/2023-09-07-wg-agenda.md#%E2%84%B9%EF%B8%8F-spec-writing-with-bikeshed-using-the-latest-webidl-and-infra-standard-conventions>ℹ️ Spec writing with Bikeshed using the latest WebIDL and Infra standard conventions

Big PR was merged and established a modernized foundation to define our new features on. Joshua Bell will give an update on spec writing using the latest conventions in Web IDL and Infra and shares his Bikeshed tips that will make editors’ and other spec contributors' work easier.

Background material:

  *   Writing Procedural Specs<https://garykac.github.io/procspec/>
  *   Writing Specifications with Bikeshed<https://dlaliberte.github.io/bikeshed-intro/>
  *   Bikeshed Docs<https://speced.github.io/bikeshed/>

<https://github.com/webmachinelearning/meetings/blob/main/telcons/2023-09-07-wg-agenda.md#%E2%84%B9%EF%B8%8F-enhancements>ℹ️ Enhancements

  *   split() into sizes not widely supported
     *    issue: webmachinelearning/webnn#392<https://github.com/webmachinelearning/webnn/issues/392>
  *   Clarify the restriction for minValue and maxValue of MLClampOptions
     *    issue: webmachinelearning/webnn#396<https://github.com/webmachinelearning/webnn/issues/396>
  *   Type of parameters should match the input data type
     *    issue: webmachinelearning/webnn#442<https://github.com/webmachinelearning/webnn/issues/442>

<https://github.com/webmachinelearning/meetings/blob/main/telcons/2023-09-07-wg-agenda.md#%E2%84%B9%EF%B8%8F-security-considerations>ℹ️ Security considerations

  *   Computation control-flow attack based on weights / constants change
     *    issue: webmachinelearning/webnn#443<https://github.com/webmachinelearning/webnn/issues/443>

<https://github.com/webmachinelearning/meetings/blob/main/telcons/2023-09-07-wg-agenda.md#%E2%84%B9%EF%B8%8F-new-features>ℹ️ New features

  *   Allow to hint number of threads for CPU MLContext
     *    issue: webmachinelearning/webnn#436<https://github.com/webmachinelearning/webnn/issues/436>