<div dir="ltr"><div class="inbox-inbox-uyb8Gf"><div><div class="inbox-inbox-F3hlO"><div dir="ltr"><div class="inbox-inbox-m_491356864120994633inbox-inbox-i3"><div dir="ltr">** Submission deadline is extended to Friday, October 12, 2018 (Anywhere on Earth) **<br><br>****************************************************************************************************<br>> Infer to Control: Probabilistic Reinforcement Learning and Structured Control<br>> NIPS 2018 Workshop<br>> Saturday, December 8<br>> Montréal, Canada<br>> Website: <a href="https://sites.google.com/view/infer2control-nips2018" target="_blank">https://sites.google.com/view/infer2control-nips2018</a> <br>> Please address questions to: <a href="mailto:infer2control.nips2018@gmail.com" target="_blank">infer2control.nips2018@gmail.com</a> <br>****************************************************************************************************<br><br>Reinforcement
learning and imitation learning are effective paradigms for learning
controllers of dynamical systems from experience. These fields have been
empowered by recent success in deep learning of differentiable
parametric models, allowing end-to-end training of highly nonlinear
controllers that encompass perception, memory, prediction, and decision
making. The aptitude of these models to represent latent dynamics,
high-level goals, and long-term outcomes is unfortunately curbed by the
poor sample complexity of many current algorithms for learning these
models from experience.<br><br>Probabilistic reinforcement learning and
inference of control structure are emerging as promising approaches for
avoiding prohibitive amounts of controller–system interactions. These
methods leverage informative priors on useful behavior, as well as
controller structure, such as hierarchy and modularity, as useful
inductive biases that reduce the effective size of policy search space
and shape the optimization landscape. Intrinsic and self-supervised
signals can further guide the training process of distinct internal
components—such as perceptual embeddings, predictive models, exploration
policies, and inter-agent communication—to break down the hard holistic
problem of control into more efficiently learnable parts.<br><br>Effective
inference methods are crucial for probabilistic approaches to
reinforcement learning and structured control. Approximate control and
model-free reinforcement learning exploit latent system structure and
priors on policy structure, that are not directly evident in the
controller–system interactions, and must be inferred by the learning
algorithm. The growing interest of the reinforcement learning and
optimal control community in the application of inference methods is
synchronized well with the development by the probabilistic learning
community of powerful inference techniques, such as probabilistic
programming, variational inference, Gaussian processes, and
nonparametric regression.<br><br>This workshop is a venue for the
inference and reinforcement learning communities to come together in
discussing recent advances, developing insights, and future potential in
inference methods and their application to probabilistic reinforcement
learning and structured control. The goal of this workshop is to
catalyze tighter collaboration within and between the communities, that
will be leveraged in upcoming years to rise to the challenges of
real-world control problems.<br><br>#### IMPORTANT DATES: ####<br>- Submission deadline (extended): Friday, October 12, 2018 (Anywhere on Earth)<br>- Author notification: Monday, October 22, 2018<br>- Final version deadline: Friday, November 30, 2018<br>- Workshop: Saturday, December 8, 2018<br><br>#### SUBMISSION DETAILS: ####<br>-
Research papers are solicited on inference for reinforcement learning
and control, its theory and applications, and related fields.<br>- Contributed papers may include novel research, preliminary results, or surveys of recent results.<br>- Papers are limited to 4 pages, excluding references and appendices of any length, in the NIPS style: <a href="https://nips.cc/Conferences/2018/PaperInformation/StyleFiles" target="_blank">https://nips.cc/Conferences/2018/PaperInformation/StyleFiles</a>.<br>- Submissions must be anonymized for double-blind review.<br>-
All accepted papers will be presented as spotlights and posters, and
made publicly available as a non-archival report, allowing future
submissions to archival conferences or journals.<br>- Authors of top accepted papers will be invited to give short contributed talks.<br>- Submission link: <a href="https://cmt3.research.microsoft.com/INFER2CONTROL2018" target="_blank">https://cmt3.research.microsoft.com/INFER2CONTROL2018</a><br>- Please check the workshop website for the latest updates: <a href="https://sites.google.com/view/infer2control-nips2018" target="_blank">https://sites.google.com/view/infer2control-nips2018</a><br><br>#### ORGANIZERS: ####<br>- Leslie Kaelbling<br>- Martin Riedmiller<br>- Marc Toussaint<br>- Igor Mordatch</div></div></div></div></div></div><div class="inbox-inbox-uyb8Gf"><div><div class="inbox-inbox-i3"><div dir="ltr"><div class="inbox-inbox-m_491356864120994633inbox-inbox-i3"><div dir="ltr">- Roy Fox<br>- Tuomas Haarnoja</div></div></div></div></div></div></div>