I n this paper, we propose new methods to approximate probability distributions that are incompletely specified. We compare these methods to the use of maximum entropy and quantify the accuracy of all methods within the context of an illustrative example. We show that within the context of our example, the methods we propose are more accurate than existing methods.

Lattice QCD studies on charmonium at finite temperature are presented After a discussion about problems for the Maximum Entropy Method applied to finite temperature lattice QCD, I show several results on charmonium spectral functions. The “wave function” of charmonium is also discussed to study the spatial correlation between quark and anti-quark in deconfinement phase.

The maximum entropy (ME) model of a knowledge base R consisting of relational probabilistic conditionals can be defined referring to the set of all ground instances of the conditionals. The logic FO-PCL employs the notion of parametric uniformity for avoiding the full grounding of R. We present an implementation of a rule system transforming R into a knowledge base that is parametrically unifor...

We consider the problem of choosing the ridge parameter. Two penalized maximum likelihood (PML) criteria based on a distribution-free and a datadependent penalty function are proposed. These PML criteria can be considered as “continuous” versions of AIC. A systematic simulation is conducted to compare the suggested criteria to several existing methods. The simulation results strongly support th...

This paper explores the problem of detecting sentence-level forum authority claims in online discussions. Using a maximum entropy model, we explore a variety of strategies for extracting lexical features in a sparse training scenario, comparing knowledgeand datadriven methods (and combinations). The augmentation of lexical features with parse context is also investigated. We find that certain m...

The well known maximum-entropy principle due to Jaynes, which states that given mean parameters, the maximum entropy distribution matching them is in an exponential family has been very popular in machine learning due to its “Occam’s razor” interpretation. Unfortunately, calculating the potentials in the maximumentropy distribution is intractable [BGS14]. We provide computationally efficient ve...

Deep learning achieves remarkable generalization capability with overwhelming number of model parameters. Theoretical understanding of deep learning generalization receives recent attention yet remains not fully explored. This paper attempts to provide an alternative understanding from the perspective of maximum entropy. We first derive two feature conditions that softmax regression strictly ap...

Given a fuzzy variable, what is the degree of difficulty of predicting the specified value that the fuzzy variable will take? Entropy provides a measure to characterize the fuzziness. Given some constraints, there are usually multiple compatible membership functions. Which membership function shall we take? The maximum entropy principle attempts to select the membership function that maximizes ...

For each transition probability, MEMM uses maximum entropy model. Maximum entropy model is used to model the data distribution, which gives us as much information as possible. Without any constraints, the uniform distribution gives us maximum entropy. However, we have training data which gives some facts about the true distribution. What maximum entropy model does is to maximize the entropy of ...

Based on the Markov model and the basic theory of information entropy, this paper puts forward a new method for optimizing the location of observation points in order to obtain more information from limited geological investigation. According to the existing data from observation points data, classification of tunnel geological lithology was performed, and various lithology distribution were de...