Results 1  10
of
14
Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation
"... Accurate modeling of geographic distributions of species is crucial to various applications in ecology and conservation. The best performing techniques often require some parameter tuning, which may be prohibitively timeconsuming to do separately for each species, or unreliable for small or biased ..."
Abstract

Cited by 131 (2 self)
 Add to MetaCart
Accurate modeling of geographic distributions of species is crucial to various applications in ecology and conservation. The best performing techniques often require some parameter tuning, which may be prohibitively timeconsuming to do separately for each species, or unreliable for small or biased datasets. Additionally, even with the abundance of good quality data, users interested in the application of species models need not have the statistical knowledge required for detailed tuning. In such cases, it is desirable to use ‘‘default settings’’, tuned and validated on diverse datasets. Maxent is a recently introduced modeling technique, achieving high predictive accuracy and enjoying several additional attractive properties. The performance of Maxent is influenced by a moderate number of parameters. The first contribution of this paper is the empirical tuning of these parameters. Since many datasets lack information about species absence, we present a tuning method that uses presenceonly data. We evaluate our method on independently collected highquality presenceabsence data. In addition to tuning, we introduce several concepts that improve the predictive accuracy and running time of Maxent. We introduce ‘‘hinge features’ ’ that model more complex relationships in the training data; we describe a new logistic output format that gives an estimate of probability of presence; finally we explore ‘‘background sampling’’ strategies that cope with sample selection bias and decrease modelbuilding time. Our evaluation, based on a diverse dataset of 226 species from 6 regions, shows: 1) default settings tuned on presenceonly data achieve performance which is almost as good as if they had been tuned on the evaluation data itself; 2) hinge features substantially improve model
A tutorial introduction to the minimum description length principle
 in Advances in Minimum Description Length: Theory and Applications. 2005
"... ..."
(Show Context)
When Ignorance is Bliss
 UAI 2004
, 2004
"... It is commonlyaccepted wisdom that more information is better, and that information should never be ignored. Here we argue, using both a Bayesian and a nonBayesian analysis, that in some situations you are better off ignoring information if your uncertainty is represented by a set of probability m ..."
Abstract

Cited by 13 (4 self)
 Add to MetaCart
It is commonlyaccepted wisdom that more information is better, and that information should never be ignored. Here we argue, using both a Bayesian and a nonBayesian analysis, that in some situations you are better off ignoring information if your uncertainty is represented by a set of probability measures. These include situations in which the information is relevant for the prediction task at hand. In the nonBayesian analysis, we show how ignoring information avoids dilation, the phenomenon that additional pieces of information sometimes lead to an increase in uncertainty. In the Bayesian analysis, we show that for small sample sizes and certain prediction tasks, the Bayesian posterior based on a noninformative prior yields worse predictions than simply ignoring the given information.
Foundations for Bayesian networks
, 2001
"... Bayesian networks are normally given one of two types of foundations: they are either treated purely formally as an abstract way of representing probability functions, or they are interpreted, with some causal interpretation given to the graph in a network and some standard interpretation of probabi ..."
Abstract

Cited by 13 (8 self)
 Add to MetaCart
Bayesian networks are normally given one of two types of foundations: they are either treated purely formally as an abstract way of representing probability functions, or they are interpreted, with some causal interpretation given to the graph in a network and some standard interpretation of probability given to the probabilities specified in the network. In this chapter I argue that current foundations are problematic, and put forward new foundations which involve aspects of both the interpreted and the formal approaches. One standard approach is to interpret a Bayesian network objectively: the graph in a Bayesian network represents causality in the world and the specified probabilities are objective, empirical probabilities. Such an interpretation founders when the Bayesian network independence assumption (often called the causal Markov condition) fails to hold. In §2 I catalogue the occasions when the independence assumption fails, and show that such failures are pervasive. Next, in §3, I show that even where the independence assumption does hold objectively, an agent’s causal knowledge is unlikely to satisfy the assumption with respect to her subjective probabilities, and that slight differences between an agent’s subjective Bayesian network and an objective Bayesian network can lead to large differences between probability distributions determined by these networks. To overcome these difficulties I put forward logical Bayesian foundations in §5. I show that if the graph and probability specification in a Bayesian network are thought of as an agent’s background knowledge, then the agent is most rational if she adopts the probability distribution determined by the
Strong Entropy Concentration, Game Theory and Algorithmic Randomness
, 2001
"... . We give a characterization of Maximum Entropy/Minimum Relative Entropy inference by providing two `strong entropy concentration ' theorems. These theorems unify and generalize Jaynes' `concentration phenomenon' and Van Campenhout and Cover's `conditional limit theorem' ..."
Abstract

Cited by 4 (2 self)
 Add to MetaCart
. We give a characterization of Maximum Entropy/Minimum Relative Entropy inference by providing two `strong entropy concentration ' theorems. These theorems unify and generalize Jaynes' `concentration phenomenon' and Van Campenhout and Cover's `conditional limit theorem'. The theorems characterize exactly in what sense a `prior' distribution Q conditioned on a given constraint and the distribution ~ P minimizing D(P jjQ) over all P satisfying the constraint are `close' to each other. We show how our theorems are related to `universal models ' for exponential families, thereby establishing a link with Rissanen's MDL/stochastic complexity. We then apply our theorems to establish the relationship (A) between entropy concentration and a gametheoretic characterization of Maximum Entropy Inference due to Topse and others; (B) between maximum entropy distributions and sequences that are random (in the sense of MartinLof/Kolmogorov) with respect to the given constraint. These two applications have strong implications for the use of Maximum Entropy distributions in sequential prediction tasks, both for the logarithmic loss and for general loss functions. We identify circumstances under which Maximum Entropy predictions are almost optimal. 1
Strong Entropy Concentration, Coding, Game Theory and Randomness
, 2001
"... We give a characterization of Maximum Entropy/Minimum Relative Entropy inference by providing two `strong entropy concentration' theorems. These theorems unify and generalize Jaynes' `concentration phenomenon' and Van Campenhout and Cover's `conditional limit theorem'. The t ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
We give a characterization of Maximum Entropy/Minimum Relative Entropy inference by providing two `strong entropy concentration' theorems. These theorems unify and generalize Jaynes' `concentration phenomenon' and Van Campenhout and Cover's `conditional limit theorem'. The theorems characterize exactly in what sense a `prior' distribution Q conditioned on a given constraint and the distribution ~ P minimizing D(PQ) over all P satisfying the constraint are `close' to each other. We show how our theorems are related to `universal models' for exponential families, thereby establishing a link with Rissanen's MDL/stochastic complexity. We then apply our theorems to establish the relationship (A) between entropy concentration and a gametheoretic characterization of Maximum Entropy Inference due to Topse and others; (B) between maximum entropy distributions and sequences that are random (in the sense of MartinLöf/Kolmogorov) with respect to the given constraint. These two applications have strong implications for the use of Maximum Entropy distributions in sequential prediction tasks, both for the logarithmic loss and for general loss functions. We identify circumstances under which Maximum Entropy predictions are almost optimal.
In Defence of Objective Bayesianism
"... Objective Bayesianism purports to tell us how strongly it is rational for an agent to believe in the propositions that can be expressed in his language. The degree of belief an agent has in a proposition can be represented by a number in the interval [0, 1] with 0 indicating that the agent is certai ..."
Abstract
 Add to MetaCart
Objective Bayesianism purports to tell us how strongly it is rational for an agent to believe in the propositions that can be expressed in his language. The degree of belief an agent has in a proposition can be represented by a number in the interval [0, 1] with 0 indicating that the agent is certain that the proposition is false and 1 indicating
Improvements on KAMETtoBayesianNetwork Transformations?
"... Abstract. KAMET is a modelbased methodology designed to manage knowledge acquisition from multiple knowledge sources that leads toagraphicalmodelthatrepresentscausal relations [2]. In the past, all the inference methods developed for KAMET were rulebased; and thus, produced a loss of the uncertain ..."
Abstract
 Add to MetaCart
Abstract. KAMET is a modelbased methodology designed to manage knowledge acquisition from multiple knowledge sources that leads toagraphicalmodelthatrepresentscausal relations [2]. In the past, all the inference methods developed for KAMET were rulebased; and thus, produced a loss of the uncertainty information included in the models. In [3] two transformations are presented, which allow the use of Bayesian networks as an inference engine for KAMET models that include probabilistic uncertainty. In this paper we present a simple proof of the equivalence between those transformations in the sense of yielding the same results under equivalent evidence, as well as some improvements made to the KAMET Modelling Language and the definition of the transformations (where an error is corrected). 1
unknown title
"... It is commonlyaccepted wisdom that more information is better, and that information should never be ignored. Here we argue, using both a Bayesian and a nonBayesian analysis, that in some situations you are better off ignoring information if your uncertainty is represented by a set of probability m ..."
Abstract
 Add to MetaCart
(Show Context)
It is commonlyaccepted wisdom that more information is better, and that information should never be ignored. Here we argue, using both a Bayesian and a nonBayesian analysis, that in some situations you are better off ignoring information if your uncertainty is represented by a set of probability measures. These include situations in which the information is relevant for the prediction task at hand. In the nonBayesian analysis, we show how ignoring information avoids dilation, the phenomenon that additional pieces of information sometimes lead to an increase in uncertainty. In the Bayesian analysis, we show that for small sample sizes and certain prediction tasks, the Bayesian posterior based on a noninformative prior yields worse predictions than simply ignoring the given information. 1
Taking the Sting out of Subjective Probability
, 2001
"... Introduction In the Bayesian approach to reasoning under uncertainty, probability is viewed as a subjective notion. While the Bayesian approach has strong theoretical foundations (see, e.g. De Finetti (1974); Savage (1954)) and typically performs well in practice, several aspects of the interpretat ..."
Abstract
 Add to MetaCart
Introduction In the Bayesian approach to reasoning under uncertainty, probability is viewed as a subjective notion. While the Bayesian approach has strong theoretical foundations (see, e.g. De Finetti (1974); Savage (1954)) and typically performs well in practice, several aspects of the interpretation of subjective probability remain controversial and problematic. In order to deal with such concerns, one may attempt to define the meaning of subjective probabilities in nonprobabilistic terms. In this paper, we give such a `meaning'. Our proposal differs markedly from existing ones in that it explicitly accounts for subjective probabilities that lead to good predictions of only some, not all aspects of a domain. In this way we manage to avoid some of the problematic aspects of (the traditional interpretations of) subjective probability. Specifically, we argue that traditional use of subjective probability sometimes leads to overly