RECORD
Smoothed Hierarchical Dirichlet Process: A Non-Parametric Approach to Constraint Measures
Machine Learning · Submitted 16 Apr 2016
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim Our experiment using NIPS keywords has shown the desirable effect of the model.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Time-varying mixture densities occur in many scenarios, for example, the distributions of keywords that appear in publications may evolve from year to year, video frame features associated with multiple targets may evolve in a sequence. Any models that realistically cater to this phenomenon must exhibit two important properties: the underlying mixture densities must have an unknown number of mixtures, and there must be some "smoothness" constraints in place for the adjacent mixture densities. The traditional Hierarchical Dirichlet Process (HDP) may be suited to the first property, but certainly not the second. This is due to how each random measure in the lower hierarchies is sampled independent of each other and hence does not facilitate any temporal correlations. To overcome such shortcomings, we proposed a new Smoothed Hierarchical Dirichlet Process (sHDP). The key novelty of this model is that we place a temporal constraint amongst the nearby discrete measures $\{G_j\}$ in the form of symmetric Kullback-Leibler (KL) Divergence with a fixed bound $B$. Although the constraint we place only involves a single scalar value, it nonetheless allows for flexibility in the corresponding successive measures. Remarkably, it also led us to infer the model within the stick-breaking process where the traditional Beta distribution used in stick-breaking is now replaced by a new constraint calculated from $B$. We present the inference algorithm and elaborate on its solutions. Our experiment using NIPS keywords has shown the desirable effect of the model.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Time-varying mixture densities occur in many scenarios, for example, the distributions of keywords that appear in publications may evolve from year to year, video frame features associated with multiple targets may evolve in a sequence. | background 1.00 | 0.06 | 0.02 | hedged possibility 0.76 |
| s1 | Any models that realistically cater to this phenomenon must exhibit two important properties: the underlying mixture densities must have an unknown number of mixtures, and there must be some "smoothness" constraints in place for the adjacent mixture densities. | background 0.92 | 0.06 | 0.06 | asserted 0.87 |
| s2 | The traditional Hierarchical Dirichlet Process (HDP) may be suited to the first property, but certainly not the second. | gap 0.91 | 0.09 | 0.06 | hedged possibility 0.66 |
| s3 | This is due to how each random measure in the lower hierarchies is sampled independent of each other and hence does not facilitate any temporal correlations. | gap 0.65 | 0.12 | 0.03 | asserted 0.99 |
| s4 | To overcome such shortcomings, we proposed a new Smoothed Hierarchical Dirichlet Process (sHDP). | contribution 0.99 | 0.90 | 0.11 | asserted 1.00 |
| s5 | The key novelty of this model is that we place a temporal constraint amongst the nearby discrete measures $\{G_j\}$ in the form of symmetric Kullback-Leibler (KL) Divergence with a fixed bound $B$. | method 0.56 | 0.92 | 0.23 | asserted 1.00 |
| s6 | Although the constraint we place only involves a single scalar value, it nonetheless allows for flexibility in the corresponding successive measures. | interpretation 0.59 | 0.77 | 0.09 | asserted 0.98 |
| s7 | Remarkably, it also led us to infer the model within the stick-breaking process where the traditional Beta distribution used in stick-breaking is now replaced by a new constraint calculated from $B$. | method 0.78 | 0.94 | 0.09 | asserted 1.00 |
| s8 | We present the inference algorithm and elaborate on its solutions. | method 0.67 | 0.90 | 0.04 | asserted 0.88 |
| s9 | Our experiment using NIPS keywords has shown the desirable effect of the model. | result 0.99 | 0.96 | 0.37 | asserted 0.99 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Time-varying mixture densities occur in many scenarios, for example, the distributions of keywords that appear in publications may evolve from year to year, video frame features associated with multiple targets may evolve in a sequence. Any models that realistically cater to this phenomenon must exhibit two important properties: the underlying mixture densities must have an unknown number of mixtures, and there must be some "smoothness" constraints in place for the adjacent mixture densities. The traditional Hierarchical Dirichlet Process (HDP) may be suited to the first property, but certainly not the second. This is due to how each random measure in the lower hierarchies is sampled independent of each other and hence does not facilitate any temporal correlations. To overcome such shortcomings, we proposed a new Smoothed Hierarchical Dirichlet Process (sHDP). The key novelty of this model is that we place a temporal constraint amongst the nearby discrete measures $\{G_j\}$ in the form of symmetric Kullback-Leibler (KL) Divergence with a fixed bound $B$. Although the constraint we place only involves a single scalar value, it nonetheless allows for flexibility in the corresponding successive measures. Remarkably, it also led us to infer the model within the stick-breaking process where the traditional Beta distribution used in stick-breaking is now replaced by a new constraint calculated from $B$. We present the inference algorithm and elaborate on its solutions. Our experiment using NIPS keywords has shown the desirable effect of the model.
Human review
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.