<div dir="ltr"><div class="gmail_default" style="font-family:verdana,sans-serif"><font color="#000000">Apologies for cross-posting. </font><span style="color:rgb(0,0,0)">We are extending the submission deadline to 20th March 2020. More details below.</span></div><div class="gmail_quote"><div dir="ltr"><div style="font-family:verdana,sans-serif"><div><span style="color:rgb(0,0,0)"><span><br></span></span></div><div><span style="color:rgb(0,0,0)"><span>Deep</span> <span>Declarative</span> <span>Networks</span> CVPR 2020 Workshop -- Call for Papers</span><br></div><div class="gmail_quote" style="font-family:Arial,Helvetica,sans-serif"><div dir="ltr"><div class="gmail_quote"><div dir="ltr"><div><font color="#000000" face="verdana, sans-serif">*****************************************************</font></div><div><font color="#000000" face="verdana, sans-serif"><br></font></div><div><div><font color="#000000" face="verdana, sans-serif">Website: <a href="https://urldefense.proofpoint.com/v2/url?u=http-3A__cvpr2020.deepdeclarativenetworks.com_&d=DwMFaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=0w3solp5fswiyWF2RL6rStC00sfV_qVWqlcnmGhKrRoTMim5rLso-FTzVW2NfNyo&m=nSaJf4GGDWvvPmYROUEbLOOSoG4UJSFiai5wZ3c0UJw&s=uiU8xfHay6u2ep2H7e2L9gyRN9rykj6ORELL9gxN10o&e=" target="_blank">http://cvpr2020.deepdeclarativenetworks.com/</a></font></div><div><font color="#000000" face="verdana, sans-serif">Submission Deadline: 20th March 2020</font></div><div><font color="#000000" face="verdana, sans-serif">Workshop on 14th June 2020 in conjunction with CVPR 2020, Seattle, WA</font></div><div><font color="#000000" face="verdana, sans-serif">Contact us at </font><font face="verdana, sans-serif"><a href="mailto:cvpr2020@deepdeclarativenetworks.com" target="_blank">cvpr2020@deepdeclarativenetworks.com</a></font></div><div><font color="#000000" face="verdana, sans-serif"><br></font></div><div><font color="#000000" face="verdana, sans-serif">*****************************************************</font></div></div><div><font color="#000000" face="verdana, sans-serif"><br></font></div><div><font color="#000000" face="verdana, sans-serif">ABOUT</font></div><div><font color="#000000" face="verdana, sans-serif">******</font></div><div><font color="#000000" face="verdana, sans-serif">Conventional <span>deep</span> learning architectures involve composition of simple feedforward processing functions that are explicitly defined. Recently, researchers have been exploring <span>deep</span> learning models with implicitly defined components, where the output of the node involves solving an optimization problem in the forward pass. To distinguish these from conventional <span>deep</span> learning models, we call them <span>deep</span> <span>declarative</span> <span>networks</span>, borrowing nomenclature from the programming languages community.<br></font></div><div><font color="#000000" face="verdana, sans-serif"><br></font></div><div><font color="#000000" face="verdana, sans-serif">This workshop explores the advantages (and potential shortcomings) of <span>declarative</span> <span>networks</span> and their variants, bringing ideas developed in different contexts under the umbrella of <span>deep</span> <span>declarative</span> <span>networks</span>. We will discuss technical issues that need to be overcome in developing such models and applications of these models to computer vision problems that show benefit over conventional approaches. For details on the topics, program, and speakers, please visit </font><a href="https://urldefense.proofpoint.com/v2/url?u=http-3A__cvpr2020.deepdeclarativenetworks.com_&d=DwMFaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=0w3solp5fswiyWF2RL6rStC00sfV_qVWqlcnmGhKrRoTMim5rLso-FTzVW2NfNyo&m=nSaJf4GGDWvvPmYROUEbLOOSoG4UJSFiai5wZ3c0UJw&s=uiU8xfHay6u2ep2H7e2L9gyRN9rykj6ORELL9gxN10o&e=" target="_blank"><font face="verdana, sans-serif">http://cvpr2020.deepdeclarativenetworks.com/</font><font color="#000000" face="verdana, sans-serif"><br></font></a></div><div><font face="verdana, sans-serif"><br></font></div><div><div><font face="verdana, sans-serif">TOPICS</font></div><div><font face="verdana, sans-serif">******</font></div></div><div><font color="#000000" face="verdana, sans-serif">We invite paper submissions of up to four (4) pages describing work in related areas including: </font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">* </span><span>Declarative</span> end-to-end learnable processing nodes </font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">* </span>Differentiable constrained and unconstrained (non-convex) optimization problems </font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">*</span> Differentiable convex optimization problems and techniques </font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">*</span> Imposing hard constraints in <span>deep</span> learning models </font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">*</span> Backpropagation through statistical and physic<span class="gmail_default">s</span> models </font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">*</span> Applications of the above to problems in computer vision<span class="gmail_default">, such as</span></font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">     ** </span>differentiable rendering</font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">     ** </span>differentiable 3d models </font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">     ** </span>reinforcement learning</font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">     ** </span>action recognition</font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">     ** </span>meta-learning, etc<span class="gmail_default">.</span><br></font></div><div></div><div><font color="#000000" face="verdana, sans-serif"><br></font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">SUBMISSIONS</span></font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default">************</span></font></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default"></span>Accepted submissions will be presented as short orals or posters at the workshop and will appear on the workshop website. Authors will be given an opportunity to revise the submission before posting on the website. </font></div><div><font color="#000000" face="verdana, sans-serif"><br></font></div><div><font color="#000000" face="verdana, sans-serif">Papers should follow the CVPR formatting guidelines and emailed as a single PDF to <a href="mailto:cvpr2020@deepdeclarativenetworks.com" target="_blank">cvpr2020@deepdeclarativenetworks.com</a>, with subject line “Paper: ”. Submissions are not anonymous (and should include all authors’ names and affiliations). References are to be included in the 4-page limit.<span class="gmail_default"> We encourage submissions of preliminary or ongoing work. Accepted papers **will not appear** in the official IEEE proceedings. Relevant work that has previously been published or is to be presented at the main conference is also welcome.</span><br></font></div><div><font color="#000000" face="verdana, sans-serif"><br></font></div><div><div><font color="#000000" face="verdana, sans-serif">IMPORTANT DATES</font></div><div><font color="#000000" face="verdana, sans-serif">****************</font></div><div><font color="#000000" face="verdana, sans-serif">Submission deadline: 20 March 2020</font></div><div><font color="#000000" face="verdana, sans-serif">Author Notification: 13 April 2020</font></div><div><font color="#000000" face="verdana, sans-serif">Camera-ready deadline: 1 May 2020</font></div><div><font color="#000000" face="verdana, sans-serif">Workshop date: 14 June 2020</font></div><font color="#000000" face="verdana, sans-serif"><br></font></div><div><div><font color="#000000" face="verdana, sans-serif">Organizers: Stephen Gould<span class="gmail_default"> (ANU)</span>, Anoop Cherian<span class="gmail_default"> (MERL)</span>, Dylan Campbell<span class="gmail_default"> (ANU)</span>, and Richard Hartley<span class="gmail_default"> (ANU)</span></font></div></div><div><font color="#000000" face="verdana, sans-serif"><span class="gmail_default"><br></span></font></div><div><font color="#000000" face="verdana, sans-serif">Email: <a href="mailto:cvpr2020@deepdeclarativenetworks.com" target="_blank">cvpr2020@deepdeclarativenetworks.com</a> </font></div><div><font color="#000000" face="verdana, sans-serif">Website: <a href="https://urldefense.proofpoint.com/v2/url?u=http-3A__cvpr2020.deepdeclarativenetworks.com_&d=DwMFaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=0w3solp5fswiyWF2RL6rStC00sfV_qVWqlcnmGhKrRoTMim5rLso-FTzVW2NfNyo&m=nSaJf4GGDWvvPmYROUEbLOOSoG4UJSFiai5wZ3c0UJw&s=uiU8xfHay6u2ep2H7e2L9gyRN9rykj6ORELL9gxN10o&e=" target="_blank">http://cvpr2020.deepdeclarativenetworks.com/ </a></font></div><div><br></div><div><br></div><div><br></div><div><br></div><div><br></div></div></div></div></div></div>-- <br><div dir="ltr" data-smartmail="gmail_signature"><div dir="ltr"><div style="font-family:verdana,sans-serif">Thank you,</div><div style="font-family:verdana,sans-serif"><br></div><div style="font-family:verdana,sans-serif">Anoop Cherian, PhD</div><div style="font-family:verdana,sans-serif">Research Scientist,</div><div style="font-family:verdana,sans-serif">Mitsubishi Electric Research Labs,</div><div style="font-family:verdana,sans-serif">Cambridge, MA</div><div style="font-family:verdana,sans-serif"><br></div><div style="font-family:verdana,sans-serif">Phone: +1-313-413-4756</div><div style="font-family:verdana,sans-serif">Email: <a href="mailto:cherian@merl.com" target="_blank">cherian@merl.com</a></div><div style="font-family:verdana,sans-serif">Website: <a href="http://users.cecs.anu.edu.au/~cherian/" style="font-family:Arial,Helvetica,sans-serif" target="_blank">http://users.cecs.anu.edu.au/~cherian/</a></div><div style="font-family:verdana,sans-serif"></div></div></div></div>
</div><div><br></div><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div style="font-family:verdana,sans-serif"></div></div></div></div>