Supplementary Materials for Learning the Parameters of Determinantal Point Process Kernels
نویسندگان
چکیده
Gradient ascent and stochastic gradient ascent provide attractive approaches in learning parameters, Θ of DPP kernel L(Θ) because of their theoretical guarantees, but require knowledge of the gradient of the log-likelihood L(Θ). In the discrete DPP setting, this gradient can be computed straightforwardly and we provide examples for discrete Gaussian and polynomial kernels here. L(Θ) = T t=1 log(det(L A t (Θ))) − T log(det(L(Θ) + I)) , dL(Θ) dΘ = T t=1 tr L A t (Θ)
منابع مشابه
Learning the Parameters of Determinantal Point Process Kernels
Determinantal point processes (DPPs) are well-suited for modeling repulsion and have proven useful in many applications where diversity is desired. While DPPs have many appealing properties, such as efficient sampling, learning the parameters of a DPP is still considered a difficult problem due to the non-convex nature of the likelihood function. In this paper, we propose using Bayesian methods...
متن کاملEnsemble Kernel Learning Model for Prediction of Time Series Based on the Support Vector Regression and Meta Heuristic Search
In this paper, a method for predicting time series is presented. Time series prediction is a process which predicted future system values based on information obtained from past and present data points. Time series prediction models are widely used in various fields of engineering, economics, etc. The main purpose of using different models for time series prediction is to make the forecast with...
متن کاملAdvances in the Theory of Determinantal Point Processes
The theory of determinantal point processes has its roots in work in mathematical physics in the 1960s, but it is only in recent years that it has been developed beyond several specific examples. While there is a rich probabilistic theory, there are still many open questions in this area, and its applications to statistics and machine learning are still largely unexplored. Our contributions are...
متن کاملDifference Operators and Determinantal Point Processes
A point process is an ensemble of random locally finite point configurations in a space X. A convenient tool for dealing with point processes is provided by the correlation functions (they are similar to the moments of random variables [19]). The correlation functions are indexed by the natural numbers n = 1, 2, . . . , and the nth function depends on n variables ranging over X. A point process...
متن کاملIndividualness and Determinantal Point Processes for Pedestrian Detection: Supplementary Material
In this supplementary material, we describe experimental results which were not included in the paper due to the page limitation. Comparisons of NMS, QUBO, and the proposed method are provided in Section 1. A sensitivity analysis of parameters is reported in Section 2. Section 3 shows the effectiveness of the proposed quality and similarity term design by comparing it to other models. Fig. 1. D...
متن کامل