Content deleted Content added
fix grammar a bit |
noted apparent confusion over definition of "concentration parameter", added 'citation needed' for previous definition ~~~~ |
||
Line 4:
The larger the value of the concentration parameter, the more evenly distributed is the resulting distribution (the more it tends towards the [[uniform distribution]]). The smaller the value of the concentration parameter, the more sparsely distributed is the resulting distribution, with all but a few parameters having a probability near zero (in other words, the more it tends towards a distribution concentrated on a single point, the [[degenerate distribution]] defined by the [[Dirac delta function]]).
In the case of multivariate Dirichlet distributions, there is some confusion over how to define the concentration parameter. In the topic modelling literature, it is often defined as the sum of the individual Dirichlet parameters {{Cite conference
In the case of a Dirichlet distribution, a concentration parameter of 1 results in all sets of probabilities being equally likely, i.e. in this case the Dirichlet distribution of dimension ''k'' is equivalent to a uniform distribution over a ''k''-dimensional simplex. Note that this is ''not'' the same as what happens when the concentration parameter tends towards infinity. In the former case, all resulting distributions are equally likely (the distribution over distributions is uniform). In the latter case, only near-uniform distributions are likely (the distribution over distributions is highly peaked around the uniform distribution). Meanwhile, in the limit as the concentration parameter tends towards zero, only distributions with nearly all mass concentrated on one of their components are likely (the distribution over distributions is highly peaked around the ''k'' possible [[Dirac delta distribution]]s centered on one of the components, or in terms of the ''k''-dimensional simplex, is highly peaked at corners of the simplex).▼
| publisher = ACM
| doi = 10.1145/1553374.1553515
| isbn = 978-1-60558-516-1
| pages = 1105–1112
| last = Wallach
| first = Hanna M.
| coauthors = Iain Murray, Ruslan Salakhutdinov, David Mimno
| title = Evaluation methods for topic models
| booktitle = Proceedings of the 26th Annual International Conference on Machine Learning
| ___location = New York, NY, USA
| series = ICML '09
| accessdate = 2011-11-19
| date = 2009
| url = http://doi.acm.org/10.1145/1553374.1553515
}}, when discussing symmetric Dirichlet distributions (where the parameters are the same for all dimensions) it is often defined to be the value of the single Dirichlet parameter used in all dimensions{{Citation needed|date=November 2011}}. This second definition is larger by a factor of the dimension of the distribution.
▲
An example of where a sparse prior (concentration parameter much less than 1) is called for, consider a [[topic model]], which is used to learn the topics that are discussed in a set of documents, where each "topic" is described using a [[categorical distribution]] over a vocabulary of words. A typical vocabulary might have 100,000 words, leading to a 100,000-dimensional categorical distribution. The [[prior distribution]] for the parameters of the categorical distribution would likely be a [[symmetric Dirichlet distribution]]. However, a coherent topic might only have a few hundred words with any significant probability mass. Accordingly, a reasonable setting for the concentration parameter might be 0.01 or 0.001. With a larger vocabulary of around 1,000,000 words, an even smaller value, e.g. 0.0001, might be appropriate.
== References ==
{{reflist}}
==See also==
| |||