Herding Generalizes Diverse M-Best Solutions
ثبت نشده
چکیده
We show that the algorithm by Batra et al. (2012) to extract diverse M -solutions from a Conditional Random Field (called divMbest) takes exactly the form of a Herding procedure (Welling, 2009b), i.e. a deterministic dynamical system that produces a sequence of solutions that respect a set of observed moment constraints. This generalization enables us to invoke properties of Herding that show that divMbest enforces implausible constraints, and to consider better alternatives for these constraints. Our experiments in image segmentation demonstrate that DivMbest is outperformed by a substantial margin when Herding is configured to enforce plausible constraints.
منابع مشابه
Herding Generalizes Diverse M -Best Solutions
We show that the algorithm to extract diverse M -solutions from a Conditional Random Field (called divMbest [1]) takes exactly the form of a Herding procedure [2], i.e. a deterministic dynamical system that produces a sequence of hypotheses that respect a set of observed moment constraints. This generalization enables us to invoke properties of Herding that show that divMbest enforces implausib...
متن کاملDiverse M-best Solutions in MRFs
Current methods for computing theM most probable configurations under a probabilistic model produce solutions that tend to be very similar to the MAP solution and each other. This is often an undesirable property. In this paper we propose an algorithm for the M-Best Mode problem, which involves finding a diverse set of highly probable solutions under a discrete probabilistic model. Given a diss...
متن کاملDiverse M-Best Solutions in Markov Random Fields
Much effort has been directed at algorithms for obtaining the highest probability (MAP) configuration in probabilistic (random field) models. In many situations, one could benefit from additional highprobability solutions. Current methods for computing the M most probable configurations produce solutions that tend to be very similar to the MAP solution and each other. This is often an undesirab...
متن کاملHerding in humans.
Herding is a form of convergent social behaviour that can be broadly defined as the alignment of the thoughts or behaviours of individuals in a group (herd) through local interaction and without centralized coordination. We suggest that herding has a broad application, from intellectual fashion to mob violence; and that understanding herding is particularly pertinent in an increasingly intercon...
متن کاملSolitons And Periodic Solutions To The Generalized Zakharov-Kuznetsov Benjamin-Bona-Mahoney Equation
This paper studies the generalized version of theZakharov-Kuznetsov Benjamin-Bona-Mahoney equation. The functionalvariable method as well as the simplest equation method areapplied to obtain solitons and singular periodic solutions to theequation. There are several constraint conditions that arenaturally revealed in order for these specialized type ofsolutions to exist. The results of this pape...
متن کامل