<div dir="ltr"><div dir="ltr"><div class="gmail_default"><div class="gmail_default" style="font-size:small"><b>When</b>: Monday, September 9th from<b style="background-color:rgb(255,255,0)"> 10</b><b><span style="background-color:rgb(255,255,0)">:00am - 11:00am CT</span></b></div><div class="gmail_default"><div style="font-size:small"><b><br></b></div><div><b style="font-size:small">Where</b>: Talk will be given <b style="font-size:small"><font color="#0000ff">live, in-person</font></b> at<br> TTIC, 6045 S. Kenwood Avenue<br> 5th Floor, <b><font color="#ff0000" face="tahoma, sans-serif"><i>CORRECTION</i></font></b><font color="#ff0000" face="tahoma, sans-serif"><i>:</i></font> talk will be in <b><u><font color="#000000" style="background-color:rgb(255,255,0)" face="verdana, sans-serif">Room 529</font></u></b></div><div style="font-size:small"><br><b>Virtually</b>: via <a href="https://uchicagogroup.zoom.us/j/91743879600?pwd=8yMHWukTKFDpazQ90gpnvmzIPkyUPR.1" target="_blank"><b>Zoom</b></a> <br></div><div style="font-size:small"> </div><div style="font-size:small"><b>Who: </b> Sudarshan Babu, TTIC</div><div style="font-size:small"><br></div></div><div class="gmail_default" style="font-size:small"><div style="border-top:none;border-right:none;border-left:none;border-bottom:2.25pt solid rgb(11,118,159);padding:0in 0in 1pt"></div><p class="MsoNormal" style="margin:0in 0in 8pt;text-align:justify;line-height:15.6933px"><b><font face="arial, sans-serif"><br></font></b></p><p class="MsoNormal" style="margin:0in 0in 8pt;text-align:justify;line-height:15.6933px"><font face="arial, sans-serif"><span style="text-align:start"><b>Title:</b> Acquiring and Adapting Priors for Novel Tasks via Neural Meta-Architectures</span><br></font></p><div><div><font face="arial, sans-serif"><b>Abstract: </b>The ability to transfer knowledge from prior experiences to novel tasks stands as a pivotal capability of intelligent agents, including both humans and computational models. This foundational principle underlies the process of transfer learning, wherein pre-trained neural networks are fine-tuned to adapt to downstream tasks, demonstrating tremendous success, both in terms of task adaptation speed and performance. However there are several domains where, due lack of data building such foundational models (large models trained on internet scale data) is not a possibility – 3D vision, computational chemistry, computational immunology, medical imaging are examples. To address these challenges, our work focuses on designing architectures to enable faster and efficient acquisition of priors when large amounts of data is unavailable. In particular, we demonstrate that we can use neural memory to enable adaptation on non stationary distributions with only few samples. Then we demonstrate that hypernetwork (a network that generates another network) designs can acquire more generalizable priors than standard networks when trained with Model Agnostic Meta-Learning (MAML). Subsequently, we apply hypernetworks to 3D scene generation, demonstrating that they can acquire priors efficiently on just a handful of training scenes, thereby leading to faster text-to-3D generation. We then extend our hypernetwork framework to perform 3D segmentation on novel scenes with limited data by efficiently transferring priors from earlier viewed scenes. Finally, we propose a molecular generative pre-training task for downstream tasks such as molecular property prediction and target-aware drug generation, which are crucial tasks in computational immunology.</font></div></div><div><b style="font-family:arial,sans-serif"><br></b></div><div><b style="font-family:arial,sans-serif">Thesis Committee:</b><span style="font-family:arial,sans-serif"> Michael Maire (Advisor), </span><span style="font-family:arial,sans-serif">Greg Shakhnarovich (Co-Advisor), David McAllester, Aly Khan.</span><br></div><div><div><font face="arial, sans-serif"><br></font></div></div><div class="gmail_default"><br></div></div></div><div><div dir="ltr" class="gmail_signature"><div dir="ltr"><div><span style="font-family:arial,helvetica,sans-serif;font-size:x-small">Mary C. Marre</span><br></div><div><div><font face="arial, helvetica, sans-serif" size="1">Faculty Administrative Support</font></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6" size="1"><b>Toyota Technological Institute</b></font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6" size="1">6045 S. Kenwood Avenue, Rm 517</font></i></div><div><font size="1"><i><font face="arial, helvetica, sans-serif" color="#3d85c6">Chicago, IL 60637</font></i><br></font></div><div><font size="1"><i><font face="arial, helvetica, sans-serif" color="#3d85c6">773-834-1757</font></i></font></div><div><b><i><a href="mailto:mmarre@ttic.edu" target="_blank"><font face="arial, helvetica, sans-serif" size="1">mmarre@ttic.edu</font></a></i></b></div></div></div></div></div><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Fri, Aug 30, 2024 at 4:59 PM Mary Marre <<a href="mailto:mmarre@ttic.edu">mmarre@ttic.edu</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div><div style="font-size:small"><b>When</b>: Monday, September 9th from<b style="background-color:rgb(255,255,0)"> 10</b><b><span style="background-color:rgb(255,255,0)">:00am - 11:00am CT</span></b></div><div style="font-size:small"><div><b><br></b></div><div><b>Where</b>: Talk will be given <b><font color="#0000ff">live, in-person</font></b> at<br> TTIC, 6045 S. Kenwood Avenue<br> 5th Floor,<strike> <b><font color="#000000" style="">Room 530</font></b></strike></div><div><br><b>Virtually</b>: via <a href="https://uchicagogroup.zoom.us/j/91743879600?pwd=8yMHWukTKFDpazQ90gpnvmzIPkyUPR.1" target="_blank"><b>Zoom</b></a> <br></div><div> </div><div><b>Who: </b> Sudarshan Babu, TTIC</div><div><br></div></div><div><div style="font-size:small;border-top:none;border-right:none;border-left:none;border-bottom:2.25pt solid rgb(11,118,159);padding:0in 0in 1pt"></div><p class="MsoNormal" style="margin:0in 0in 8pt;text-align:justify;line-height:15.6933px"><b><font face="arial, sans-serif"><br></font></b></p><p class="MsoNormal" style="margin:0in 0in 8pt;text-align:justify;line-height:15.6933px"><font face="arial, sans-serif"><span style="text-align:start"><b>Title:</b> Acquiring and Adapting Priors for Novel Tasks via Neural Meta-Architectures</span><br></font></p><div><div><font face="arial, sans-serif"><b>Abstract: </b>The ability to transfer knowledge from prior experiences to novel tasks stands as a pivotal capability of intelligent agents, including both humans and computational models. This foundational principle underlies the process of transfer learning, wherein pre-trained neural networks are fine-tuned to adapt to downstream tasks, demonstrating tremendous success, both in terms of task adaptation speed and performance. However there are several domains where, due lack of data building such foundational models (large models trained on internet scale data) is not a possibility – 3D vision, computational chemistry, computational immunology, medical imaging are examples. To address these challenges, our work focuses on designing architectures to enable faster and efficient acquisition of priors when large amounts of data is unavailable. In particular, we demonstrate that we can use neural memory to enable adaptation on non stationary distributions with only few samples. Then we demonstrate that hypernetwork (a network that generates another network) designs can acquire more generalizable priors than standard networks when trained with Model Agnostic Meta-Learning (MAML). Subsequently, we apply hypernetworks to 3D scene generation, demonstrating that they can acquire priors efficiently on just a handful of training scenes, thereby leading to faster text-to-3D generation. We then extend our hypernetwork framework to perform 3D segmentation on novel scenes with limited data by efficiently transferring priors from earlier viewed scenes. Finally, we propose a molecular generative pre-training task for downstream tasks such as molecular property prediction and target-aware drug generation, which are crucial tasks in computational immunology.</font></div></div><div><b style="font-family:arial,sans-serif"><br></b></div><div><b style="font-family:arial,sans-serif">Thesis Committee:</b><span style="font-family:arial,sans-serif"> Michael Maire (Advisor), </span><span style="font-family:arial,sans-serif">Greg Shakhnarovich (Co-Advisor), David McAllester, Aly Khan.</span><br></div><div><div><font face="arial, sans-serif"><br></font></div></div><div style="font-size:small"><br></div><div style="font-size:small"><br></div><div style="font-size:small"><br></div></div></div><div><div dir="ltr" class="gmail_signature"><div dir="ltr"><div><span style="font-family:arial,helvetica,sans-serif;font-size:x-small">Mary C. Marre</span><br></div><div><div><font face="arial, helvetica, sans-serif" size="1">Faculty Administrative Support</font></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6" size="1"><b>Toyota Technological Institute</b></font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6" size="1">6045 S. Kenwood Avenue, Rm 517</font></i></div><div><font size="1"><i><font face="arial, helvetica, sans-serif" color="#3d85c6">Chicago, IL 60637</font></i><br></font></div><div><font size="1"><i><font face="arial, helvetica, sans-serif" color="#3d85c6">773-834-1757</font></i></font></div><div><b><i><a href="mailto:mmarre@ttic.edu" target="_blank"><font face="arial, helvetica, sans-serif" size="1">mmarre@ttic.edu</font></a></i></b></div></div></div></div></div></div>
</blockquote></div></div>