<div dir="ltr"><div>tl;dr: </div><div><br></div><div><span style="font-family:monospace,monospace">LAFI submission deadline Tue 6 Nov 2018 (EXTENDED)</span></div><div><br></div><div>or earlier!</div><div><br></div><div></div><div><div class="gmail_quote"><br><div dir="ltr"><div dir="ltr">LAFI 2019: Languages for Inference (formerly PPS)<br>================================================<br>Tuesday, 15 January 2019, Cascais/Lisbon, Portugal<br>A workshop affiliated with POPL 2019<br><a href="https://popl19.sigplan.org/track/lafi-2019" target="_blank">https://popl19.sigplan.org/track/lafi-2019</a><br><br>Important dates (anywhere on earth)<br><span style="font-family:monospace,monospace">-------------------------------------------------<br> LAFI submission deadline Tue 6 Nov 2018 (EXTENDED)<br> Notification Mon 3 Dec 2018<br> Early Registration Deadline Thu 10 Dec 2018<br> Workshop Tue 15 Jan 2019<br>-------------------------------------------------</span><br><br>Submission: <a href="https://lafi19.hotcrp.com/" target="_blank">https://lafi19.hotcrp.com/</a><br>Registration: <a href="https://popl19.sigplan.org/attending/Registration" target="_blank">https://popl19.sigplan.org/attending/Registration</a><br><br>Context<br>=======<br><br>Inference concerns re-calibrating program parameters based on<br>observed data, and has gained wide traction in machine learning and<br>data science. Inference can be driven by probabilistic analysis and<br>simulation, and through back-propagation and<br>differentiation. Languages for inference offer built-in support for<br>expressing probabilistic models and inference methods as programs, to<br>ease reasoning, use, and reuse. The recent rise of practical<br>implementations as well as research activity in inference-based<br>programming has renewed the need for semantics to help us share<br>insights and innovations.<br><br>This workshop aims to bring programming-language and machine-learning<br>researchers together to advance all aspects of languages for<br>inference. Topics include but are not limited to:<br><br>+ design of programming languages for inference and/or differentiable<br> programming;<br>+ inference algorithms for probabilistic programming languages,<br> including ones that incorporate automatic differentiation;<br>+ automatic differentiation algorithms for differentiable programming<br> languages;<br>+ probabilistic generative modelling and inference;<br>+ variational and differential modelling and inference;<br>+ semantics (axiomatic, operational, denotational, games, etc) and<br> types for inference and/or differentiable programming;<br>+ efficient and correct implementation;<br>+ and last but not least, applications of inference and/or<br> differentiable programming.<br><br>For a sense of the talks, posters, and blogs in past years, see<br><br>+ PPS-2018: <a href="http://conf.researchr.org/track/POPL-2018/pps-2018" target="_blank">http://conf.researchr.org/track/POPL-2018/pps-2018</a><br> blog: <a href="http://pps2018.soic.indiana.edu/" target="_blank">http://pps2018.soic.indiana.edu/</a><br><br>+ PPS-2017: <a href="http://conf.researchr.org/track/POPL-2017/pps-2017" target="_blank">http://conf.researchr.org/track/POPL-2017/pps-2017</a><br> blog: <a href="http://pps2017.soic.indiana.edu/" target="_blank">http://pps2017.soic.indiana.edu/</a>)<br><br>+ PPS-2016: <a href="http://conf.researchr.org/track/POPL-2016/pps-2016" target="_blank">http://conf.researchr.org/track/POPL-2016/pps-2016</a><br> blog: <a href="http://pps2016.soic.indiana.edu/" target="_blank">http://pps2016.soic.indiana.edu/</a>)<br><br>This year we are explicitly expanding the focus of the workshop from<br>statistical probabilistic programming to encompass differentiable<br>programming for statistical machine learning.<br><br>We expect this workshop to be informal, and our goal is to foster<br>collaboration and establish common ground. Thus, the proceedings will<br>not be a formal or archival publication, and we expect to spend only a<br>portion of the workshop day on traditional research talks.<br>Nevertheless, as a concrete basis for fruitful discussions, we call<br>for extended abstracts describing specific and ideally ongoing work on<br>probabilistic programming languages, semantics, and systems.<br><br>Submission guidelines<br>=====================<br><br>Extended abstracts are up to 2 pages in PDF format, excluding<br>references.<br><br>Please submit them by November 6(AoE) using HotCRP at:<br><a href="https://lafi19.hotcrp.com/" target="_blank">https://lafi19.hotcrp.com/</a><br><br>In line with the SIGPLAN Republication Policy:<br><br><a href="http://www.sigplan.org/Resources/Policies/Republication/" target="_blank">http://www.sigplan.org/Resources/Policies/Republication/</a><br><br>inclusion of extended abstracts in the programme is not intended to<br>preclude later formal publication.<br><br>Programme committee:<br><span style="font-family:monospace,monospace">Atılım Güneş Baydin University of Oxford Department of Engineering<br>Bart van Merriënboer University of Montreal<br>Christine Tasson University Paris Diderot<br>David Duvenaud University of Toronto<br>Jeffrey Siskind (co-chair) School of Electrical and Computer</span></div><div dir="ltr"><span style="font-family:monospace,monospace"> Engineering, Purdue University<br>Matthew Johnson Google Brain<br>Ohad Kammar (co-chair) University of Oxford Department of</span></div><div dir="ltr"><span style="font-family:monospace,monospace"> Computer Science<br>Praveen Narayanan Indiana University<br>Ryan Culpepper Czech Technical University<br>Sophia Gold Tezos<br>Steven Holtzen University of California Los Angeles<br>Tom Rainforth University of Oxford Department of Statistics</span><br><br></div></div>
</div></div></div>