w3c
diff --git a/‎_includes/talk-list1.html
+6-6 b/‎_includes/talk-list1.html
+6-6
@@ -1,7 +1,7 @@
-<div class="talks"><details class=talk><summary><a href="talks/privacy_first_approach_to_machine_learning.html"><img src="https://cjx1uopmt0m4q0667xmnrqpk.blob.core.windows.net/ckdor50iyc9lm07729dsrtpqu/thumbs/thumb-001.jpeg" alt="Watch Privacy-first approach to machine learning" width=200 class="tn"></a><a href="talks/privacy_first_approach_to_machine_learning.html">Privacy-first approach to machine learning</a><span class="summary"> by Philip Laszkowicz - 11 min</span><a>more…</a></summary><p><a href="talks/privacy_first_approach_to_machine_learning.html">11 minutes presentation</a></p><dl><dt>Speaker</dt><dd>Philip Laszkowicz</dd><dt>Abstract</dt><dd>The presentation will discuss how developers should be building modern web apps and what is missing in the existing ecosystem to make privacy-first ML possible including the challenges with WASI, modular web architecture, and localized analytics.</dd></dl></details>
-<details><summary><span>PENDING </span><a>Real-time ML Process of media in-browser</a><span class="summary"> by Bernard Aboba (Microsoft)</span><a>more…</a></summary><dl><dt>Speaker</dt><dd>Bernard Aboba (Microsoft)</dd><dt>Abstract</dt><dd>The presentation will discuss efficient processing of raw video in machine learning, highlighting the need to minimize memory copies and enable integration with WebGPU.</dd></dl></details>
-<details><summary><span>PENDING </span><a>ML hardware advancements: leveraging for performance while retaining portability</a><span class="summary"> by Cormac Brick (Intel)</span><a>more…</a></summary><dl><dt>Speaker</dt><dd>Cormac Brick (Intel)</dd><dd>Cormac leads Edge Inference IP architecture at Intel with a focus on both silicon and software. Cormac joined Intel as part of Movidius Acquisition in 2016 where he lead Machine Intelligence.</dd><dt>Abstract</dt><dd>There are a lot of interesting innovations coming in hardware technology, including hardware acceleration, weight compression, weight sparsity, low precision.</dd></dl></details>
-<details class=talk><summary><a href="talks/opportunities_and_challenges_for_tensorflow_js_and_beyond.html"><img src="https://cjx1uopmt0m4q0667xmnrqpk.blob.core.windows.net/ckdobxb1t766z0772f2e19nqo/thumbs/thumb-001.jpeg" alt="Watch Opportunities and Challenges for TensorFlow.js and beyond" width=200 class="tn"></a><a href="talks/opportunities_and_challenges_for_tensorflow_js_and_beyond.html">Opportunities and Challenges for TensorFlow.js and beyond</a><span class="summary"> by Jason Mayes (Google) - 10 min</span><a>more…</a></summary><p><a href="talks/opportunities_and_challenges_for_tensorflow_js_and_beyond.html">10 minutes presentation</a></p><dl><dt>Speaker</dt><dd>Jason Mayes (Google)</dd><dd>Developer Advocate for TensorFlow.js</dd><dt>Abstract</dt><dd>This talk will give a brief overview of TensorFlow.js, how it helps developers build ML-powered applications along with examples of work that is pushing the boundaries of the web, and discuss future directions for the web tech stack to help overcome barriers to ML in the web the TF.js community has encountered.</dd></dl></details>
-<details class=talk><summary><a href="talks/machine_learning_in_web_architecture.html"><img src="https://cjx1uopmt0m4q0667xmnrqpk.blob.core.windows.net/ckdo77c2k4srf07726abf6aps/thumbs/thumb-001.jpeg" alt="Watch Machine Learning in Web Architecture" width=200 class="tn"></a><a href="talks/machine_learning_in_web_architecture.html">Machine Learning in Web Architecture</a><span class="summary"> by Sangwhan Moon - 4 min</span><a>more…</a></summary><p><a href="talks/machine_learning_in_web_architecture.html">4 minutes presentation</a></p><dl><dt>Speaker</dt><dd>Sangwhan Moon</dd></dl></details>
-<details><summary><span>PENDING </span><a>Moving Deep Learning into Web Browser: How Far Can We Go?</a><span class="summary"> by Yun Ma</span><a>more…</a></summary><dl><dt>Speaker</dt><dd>Yun Ma</dd><dd>Yun Ma is a postdoc researcher in School of Software, Tsinghua University, China. He got his Ph.D. degree in Jun. 2017 from Peking University. His research interests lie in mobile computing, Web systems, and services computing. He has published several papers on WWW and ACM Trans on the Web. Recently he focuses on how to enable browsers to support deep learning tasks better.</dd><dt>Abstract</dt><dd>Recently, several JavaScript-based deep learning frameworks have emerged, making it possible to perform deep learning tasks directly in browsers. However, little is known on what and how well we can do with these frameworks for deep learning in browsers. In this talk, I’ll present our recent empirical study of deep learning in browsers. We survey 7 most popular JavaScript-based deep learning frameworks, investigating to what extent deep learning tasks have been supported in browsers so far. Then we measure the performance of different frameworks when running different deep learning tasks. Finally, we dig out the performance gap between deep learning in browsers and on native platforms by comparing the performance of TensorFlow.js and TensorFlow in Python. Our findings could help application developers, deep-learning framework vendors and browser vendors to improve the efficiency of deep learning in browsers. The content of this talk was published in WWW 2019.</dd></dl></details>
+<div class="talks"><details class=talk><summary><a href="talks/privacy_first_approach_to_machine_learning.html"><img src="https://cjx1uopmt0m4q0667xmnrqpk.blob.core.windows.net/ckdor50iyc9lm07729dsrtpqu/thumbs/thumb-001.jpeg" alt="Watch Privacy-first approach to machine learning" width=200 class="tn"></a><a href="talks/privacy_first_approach_to_machine_learning.html">Privacy-first approach to machine learning</a><span class="summary"> by Philip Laszkowicz - 11 min <span></span></span></summary><p><a href="talks/privacy_first_approach_to_machine_learning.html">11 minutes presentation</a></p><dl><dt>Speaker</dt><dd>Philip Laszkowicz</dd><dt>Abstract</dt><dd>The presentation will discuss how developers should be building modern web apps and what is missing in the existing ecosystem to make privacy-first ML possible including the challenges with WASI, modular web architecture, and localized analytics.</dd></dl></details>
+<details><summary><span>PENDING </span><a>Real-time ML Process of media in-browser</a><span class="summary"> by Bernard Aboba (Microsoft) <span></span></span></summary><dl><dt>Speaker</dt><dd>Bernard Aboba (Microsoft)</dd><dt>Abstract</dt><dd>The presentation will discuss efficient processing of raw video in machine learning, highlighting the need to minimize memory copies and enable integration with WebGPU.</dd></dl></details>
+<details><summary><span>PENDING </span><a>ML hardware advancements: leveraging for performance while retaining portability</a><span class="summary"> by Cormac Brick (Intel) <span></span></span></summary><dl><dt>Speaker</dt><dd>Cormac Brick (Intel)</dd><dd>Cormac leads Edge Inference IP architecture at Intel with a focus on both silicon and software. Cormac joined Intel as part of Movidius Acquisition in 2016 where he lead Machine Intelligence.</dd><dt>Abstract</dt><dd>There are a lot of interesting innovations coming in hardware technology, including hardware acceleration, weight compression, weight sparsity, low precision.</dd></dl></details>
+<details class=talk><summary><a href="talks/opportunities_and_challenges_for_tensorflow_js_and_beyond.html"><img src="https://cjx1uopmt0m4q0667xmnrqpk.blob.core.windows.net/ckdobxb1t766z0772f2e19nqo/thumbs/thumb-001.jpeg" alt="Watch Opportunities and Challenges for TensorFlow.js and beyond" width=200 class="tn"></a><a href="talks/opportunities_and_challenges_for_tensorflow_js_and_beyond.html">Opportunities and Challenges for TensorFlow.js and beyond</a><span class="summary"> by Jason Mayes (Google) - 10 min <span></span></span></summary><p><a href="talks/opportunities_and_challenges_for_tensorflow_js_and_beyond.html">10 minutes presentation</a></p><dl><dt>Speaker</dt><dd>Jason Mayes (Google)</dd><dd>Developer Advocate for TensorFlow.js</dd><dt>Abstract</dt><dd>This talk will give a brief overview of TensorFlow.js, how it helps developers build ML-powered applications along with examples of work that is pushing the boundaries of the web, and discuss future directions for the web tech stack to help overcome barriers to ML in the web the TF.js community has encountered.</dd></dl></details>
+<details class=talk><summary><a href="talks/machine_learning_in_web_architecture.html"><img src="https://cjx1uopmt0m4q0667xmnrqpk.blob.core.windows.net/ckdo77c2k4srf07726abf6aps/thumbs/thumb-001.jpeg" alt="Watch Machine Learning in Web Architecture" width=200 class="tn"></a><a href="talks/machine_learning_in_web_architecture.html">Machine Learning in Web Architecture</a><span class="summary"> by Sangwhan Moon - 4 min <span></span></span></summary><p><a href="talks/machine_learning_in_web_architecture.html">4 minutes presentation</a></p><dl><dt>Speaker</dt><dd>Sangwhan Moon</dd></dl></details>
+<details><summary><span>PENDING </span><a>Moving Deep Learning into Web Browser: How Far Can We Go?</a><span class="summary"> by Yun Ma <span></span></span></summary><dl><dt>Speaker</dt><dd>Yun Ma</dd><dd>Yun Ma is a postdoc researcher in School of Software, Tsinghua University, China. He got his Ph.D. degree in Jun. 2017 from Peking University. His research interests lie in mobile computing, Web systems, and services computing. He has published several papers on WWW and ACM Trans on the Web. Recently he focuses on how to enable browsers to support deep learning tasks better.</dd><dt>Abstract</dt><dd>Recently, several JavaScript-based deep learning frameworks have emerged, making it possible to perform deep learning tasks directly in browsers. However, little is known on what and how well we can do with these frameworks for deep learning in browsers. In this talk, I’ll present our recent empirical study of deep learning in browsers. We survey 7 most popular JavaScript-based deep learning frameworks, investigating to what extent deep learning tasks have been supported in browsers so far. Then we measure the performance of different frameworks when running different deep learning tasks. Finally, we dig out the performance gap between deep learning in browsers and on native platforms by comparing the performance of TensorFlow.js and TensorFlow in Python. Our findings could help application developers, deep-learning framework vendors and browser vendors to improve the efficiency of deep learning in browsers. The content of this talk was published in WWW 2019.</dd></dl></details>
 </dl>