SPTLI 2013 - International workshop on Structured Prediction: Tractability, Learning, and Inference

Structured Prediction: Tractability, Learning, and Inference
CVPR 2013 Workshop
June 24th, 2013, Portland, Oregon
URL: http://cvpr13ws.is.tue.mpg.de/
Abstract
--------
Computer vision tasks are routinely addressed by building a statistical model
that can be learned from annotated data and then be used to perform inference
on novel test data. Rich models that express the real world faithfully most
often are intractable in that inference and estimation within the model class
are hard problems. On the other hand, scalable and efficient discriminative
models are based on simplifying assumptions that ignore important physical
constraints; as result, for many tasks high performance can only be achieved
with extremely large training sets. Therefore, a key problem in computer
vision applications is in constructing models expressive enough to solve the
task at hand while remaining tractable. Many successful examples where this
has been achieved have led to breakthroughs for computer vision, such as for
example graphcut-based image segmentation or deformable part models (DPM).
Objectives
----------
The goal of this workshop is to bring together researchers from the computer
vision and machine learning community to discuss all issues related to
tractable structured prediction models. In particular,
* Model representations
* Inference
* Estimation
In all aspects the computer vision community has been at the forefront of
developing new ideas, in representation (e.g. perturb-and-MAP, sum-product
networks, dense random fields, deep generative models), inference (e.g. dense
mean field inference, higher-order factors), and estimation (e.g. direct loss
minimization).
The theme of the workshop continues a series of recent related workshops:
* NIPS 2012 workshop on Perturbations, Optimization, and Statistics
* ICML 2012 workshop on Inferning
* NIPS 2010 Coarse2Fine workshop
Important Dates
---------------
* Deadline for submission of papers: Friday, April 26th 2013
* Notification of acceptance: Monday, May 20th, 2013
(Note: the CVPR early registration deadline is May 24th, 2013.)
* Final version of submission: Friday, May 31st, 2013
* Workshop date: Monday, June 24th, 2013
Call for Participation
----------------------
We invite authors to submit work on the topic of structured prediction in
general, including but not limited to the following:
* Model representation for structured prediction
- Probabilistic graphical models
- Energy minimization approaches
- Perturb-and-MAP, inference machines, and algorithmic representations
- Tractable subclasses of structured prediction models
- Higher-order factors
- Latent variable models
- Structure learning
- Discrete/continuous models
* Inference
- Message passing and variational methods
- Decomposition methods
- Monte Carlo methods for structured models
* Estimation and Learning
- Likelihood based learning with approximate inference
- Empirical risk minimization in structured models
- Non-decomposable loss functions
- Reinforcement learning approaches to structured prediction
Submission Instructions
-----------------------
Details regarding the submission process are available at the workshop
homepage (style files, page limits, etc.).
Please note that at least one author of each accepted paper must be available
to present the paper at the workshop. The papers accepted to the workshop
will be published at the workshop homepage and not be part of IEEE
Proceedings. We make this choice in order to allow work to be presented that
in the future can be submitted to a conference or journal.
Schedule
--------
The workshop is held as a one-day event with a morning and afternoon session.
In addition to a lunch break, long coffee breaks will be offered both in the
morning and afternoon.
Invited Speakers
----------------
* Justin Domke -- NICTA
* Pedro Felzenszwalb -- Brown University
* Danny Tarlow -- Microsoft Research
* Raquel Urtasun -- Toyota Technological Institute at Chicago
Workshop Organizers
-------------------
* Sebastian Nowozin
Microsoft Research, Cambridge, UK
* Peter Gehler
Max Planck Institute for Intelligent Systems, Tuebingen, Germany