[Lex Computer & Tech Group/LCTG] Last Chance, Register Now: Live Webinar on The Principles of Deep Learning Theory

Wed Mar 29 14:18:57 PDT 2023

FYI, for those who might be interested enough to sign up.

    – Mike Alexander

On Wednesday, March 29, 2023, 1:30 PM, Physics Today Webinars <emails at physicstoday-info.org> wrote:

Last Chance, Register Now: Live Webinar on The Principles of Deep Learning Theory<!-- at media only screen and ()---->#yiv9978411061 body, #yiv9978411061 html{Margin:0!important;padding:0!important;width:100%!important;}#yiv9978411061 h1, #yiv9978411061 h2, #yiv9978411061 h3, #yiv9978411061 h4{Margin:0;}#yiv9978411061 *{}#yiv9978411061 div .yiv9978411061filtered99999 {margin:0!important;}#yiv9978411061 table, #yiv9978411061 td{}#yiv9978411061 table{border-spacing:0!important;border-collapse:collapse!important;table-layout:auto!important;}#yiv9978411061 img{}#yiv9978411061 .yiv9978411061yshortcuts a{border-bottom:none!important;}#yiv9978411061 .yiv9978411061mobile-link--footer a, #yiv9978411061 a .yiv9978411061filtered99999 {color:inherit!important;text-decoration:underline!important;}#yiv9978411061 .yiv9978411061email-width, #yiv9978411061 .yiv9978411061row{Margin:0 auto!important;}@media screen and (max-width:504px){#yiv9978411061 .yiv9978411061row .yiv9978411061stack-column{display:block!important;width:100%!important;max-width:100%!important;direction:ltr!important;min-width:100%!important;}#yiv9978411061 body .yiv9978411061ee-show-on-desktop{display:none!important;}#yiv9978411061 table .yiv9978411061ee-hide-on-desktop{display:table;max-height:none;visibility:visible;width:100%!important;}#yiv9978411061 .yiv9978411061email-width{width:100%!important;}}@media only screen and (min-width:505px){#yiv9978411061 .yiv9978411061row .yiv9978411061stack-column{min-width:0!important;}#yiv9978411061 .yiv9978411061ee-show-on-desktop{display:block!important;}#yiv9978411061 table.yiv9978411061ee-show-on-desktop{display:table!important;}#yiv9978411061 .yiv9978411061ee_columns.yiv9978411061ee-hide-on-desktop, #yiv9978411061 .yiv9978411061ee_element.yiv9978411061ee-hide-on-desktop, #yiv9978411061 .yiv9978411061ee_element.yiv9978411061x_ee-hide-on-desktop{display:none!important;}}#yiv9978411061 .yiv9978411061filtered99999 .yiv9978411061no-stack-column, #yiv9978411061 .yiv9978411061filtered99999 .yiv9978411061stack-column{Margin:0;float:none;display:inline-block!important;}#yiv9978411061 .yiv9978411061stack-column{min-width:0!important;}#yiv9978411061 .yiv9978411061no-stack-column{min-height:1px;}#yiv9978411061 body .yiv9978411061stack-column{min-width:100%!important;}#yiv9978411061 .yiv9978411061filtered99999 .yiv9978411061ee-show-on-desktop{display:block!important;}#yiv9978411061 .yiv9978411061filtered99999 table.yiv9978411061ee-show-on-desktop{display:table!important;}#yiv9978411061 .yiv9978411061ee-hide-on-desktop{visibility:hidden;width:0!important;max-height:0;display:block;}#yiv9978411061 body .yiv9978411061ee-hide-on-desktop{visibility:visible;width:100%!important;display:table;max-height:none;}#yiv9978411061 #yiv9978411061MessageViewBody .yiv9978411061ee-show-on-desktop{display:none!important;}#yiv9978411061 a .yiv9978411061filtered99999 {color:inherit!important;text-decoration:none!important;font-size:inherit!important;font-family:inherit!important;font-weight:inherit!important;line-height:inherit!important;}#yiv9978411061 #yiv9978411061MessageViewBody a{color:inherit;text-decoration:none;font-size:inherit;font-family:inherit;font-weight:inherit;line-height:inherit;}#yiv9978411061 u+#yiv9978411061body a{color:inherit;text-decoration:none;font-size:inherit;font-family:inherit;font-weight:inherit;line-height:inherit;}#yiv9978411061 span.yiv9978411061MsoHyperlink, #yiv9978411061 span.yiv9978411061MsoHyperlinkFollowed{color:inherit!important;text-decoration:none!important;}@media screen and (max-width:504px){#yiv9978411061 .yiv9978411061filtered99999 .yiv9978411061ee-show-on-desktop, #yiv9978411061 body .yiv9978411061ee-show-on-desktop{display:none!important;}#yiv9978411061 .yiv9978411061two-cols-per-row>tbody>tr>td>.yiv9978411061stack-column{max-width:100%!important;min-width:50%!important;width:50%!important;display:inline-block!important;}}#yiv9978411061 .yiv9978411061button-td>p>font{display:block;}<!--#yiv9978411061 .yiv9978411061ExternalClass p{MARGIN:0px;}#yiv9978411061 .yiv9978411061button-td:hover .yiv9978411061button-a {transition:100ms ease-in;background:rgba(255, 255, 255, 0.2) !important;}--><!--#yiv9978411061 p.yiv9978411061productfield {}@media only screen and ()----> 
| 
| 
| 
| 
| 
Get a sense of how deep learning goes beyond classical statistics
 |

| 
View in your browser

Forward this email
 |

 |

 |

 |

| 
| 
| 
| 
| 
|  |

|  |

| 
| 
|  |

| 
| 
|  |

|  |

 |

| 
|  |

| 
LIVE WEBINAR
 |

 |

| 
|  |

|  |

 |

 |

| 
The Principles of Deep Learning Theory

 |

|  |

| 
March 30, 2023 • 1:00 PM EDT

(Can't attend?  Recording Emailed Post-Event to all registrants)
 |

| 
| 
REGISTER NOW
 |

 |

| 
SEE BELOW FOR DETAILS

 |

|  |

|  |

| 
| 
|  |

 |

| 
| 
SPEAKER
 |

| 
dan roberts
 |

| 
Research Affiliate

MIT
 |

| 
Read Dan's full biography, here
 |

 |

 |

|  |

| 
| 
|  |

 |

| 
| 
SPEAKER
 |

| 
sho yaida
 |

| 
Research Scientist

Meta AI
 |

| 
Read Sho's full biography, here
 |

 |

 |

|  |

 |

 |

| 
SUMMARY:

Deep learning is an exciting approach to modern artificial intelligence that is based on artificial neural networks. The goal of this talk is to provide a blueprint—using tools from physics—for theoretically analyzing realistic deep neural networks.

Specifically, we'll analyze both the statistics of initialized deep networks and the training dynamics of such an ensemble when learning from data. To begin, we'll overview a particular limit at which such networks simplify drastically and become analytically tractable. In doing so, we'll see that this idealized limit is too simple to capture important aspects of the modern deep learning framework that distinguish it from classical statistical methods such as linear regression and kernel machines. To make progress,.... Keep Reading!

Learning Outcomes:

   - Receive an introduction to deep learning theory from a physicist's perspective, including background material on neural networks

   - Obtain an understanding of how modern artificial intelligence tools can be described in an effective theory framework

   - Get a sense of how deep learning goes beyond classical statistics

   - Get a flavor of how theory can impact practice
 |

|  |

|  |

| 
To access the event please register.

(Can't attend?  Recording Emailed Post-Event to all registrants)
 |

| 
| 
REGISTER NOW
 |

 |

 |

 |

| 
| 
| 
Physics Today is published by The American Institute of Physics

1 Physics Ellipse, College Park, MD 20740 USA +1 301 209 3100

© 2023 American Institute of Physics.
 |

| 
|  |  |

 |

| 
Unsubscribe  |  Email Preferences  |  Privacy Policy  |  Contact Us
 |

 |

 |

 |

 |

 |

| 
| 
| 
|  |

 |

 |

 |

 |

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.toku.us/pipermail/lctg-toku.us/attachments/20230329/5de260ea/attachment.htm>