Determinantal point processes for machine learning

Abstract

Determinantal point processes (DPPs) are elegant probabilistic models of repulsion that arise in quantum physics and random matrix theory. In contrast to traditional structured models like Markov random fields, which become intractable and hard to approximate in the presence of negative correlations, DPPs offer efficient and exact algorithms for sampling, marginalization, conditioning, and other inference tasks. We provide a gentle introduction to DPPs, focusing on the intuitions, algorithms, and extensions that are most relevant to the machine learning community, and show how DPPs can be applied to real-world applications like finding diverse sets of high-quality search results, building informative summaries by selecting diverse sentences from documents, modeling non-overlapping human poses in images or video, and automatically building timelines of important news stories.

120 pages

References in corpus (9)

• Loopy Belief Propagation for Approximate Inference: An Empirical Study
• Determinantal Processes and Independence
• Random matrices and determinantal processes
• Non-monotone submodular maximization under matroid and knapsack constraints
• Learning Mixtures of Submodular Shells with Application to Document Summarization
• Submodular Function Maximization via the Multilinear Relaxation and Contention Resolution Schemes
• Determinantal processes with number variance saturation
• Complexity and Completeness of Immanants
• Janossy Densities I. Determinantal Ensembles

Cited by in corpus (129)

• Explaining Machine Learning Classifiers through Diverse Counterfactual Explanations
• Diversity in Machine Learning
• Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Bounds
• Determinantal point process models and statistical inference : Extended version
• Data-Driven Design for Metamaterials and Multiscale Systems: A Review
• Deep Learning in Deterministic Computational Mechanics
• Applications of Near-Term Photonic Quantum Computers: Software and Algorithms
• Analysis of the hands in egocentric vision: A survey
• Assessing and Remedying Coverage for a Given Dataset
• Personalized Bundle List Recommendation
• Diverse Trajectory Forecasting with Determinantal Point Processes
• Diversity and Inclusion Metrics in Subset Selection
• Diverse Weighted Bipartite b-Matching
• Diversity Networks: Neural Network Compression Using Determinantal Point Processes
• Point Processes with Gaussian Boson Sampling
• METASET: Exploring Shape and Property Spaces for Data-Driven Metamaterials Design
• Learning from the Dark: Boosting Graph Convolutional Neural Networks with Diverse Negative Samples
• MO-PaDGAN: Reparameterizing Engineering Designs for Augmented Multi-objective Optimization
• How to be Fair and Diverse?
• Summary Transfer: Exemplar-based Subset Selection for Video Summarization
• t-METASET: Tailoring Property Bias of Large-Scale Metamaterial Datasets through Active Learning
• DeepGD: A Multi-Objective Black-Box Test Selection Approach for Deep Neural Networks
• Ab-initio study of interacting fermions at finite temperature with neural canonical transformation
• Precise Programmable Quantum Simulations with Optical Lattices
• Scalable Log Determinants for Gaussian Process Kernel Learning
• Diversified Hidden Markov Models for Sequential Labeling
• Advances in Variational Inference
• Batch Active Preference-Based Learning of Reward Functions
• Understanding Diversity in Session-Based Recommendation
• MEMe: An Accurate Maximum Entropy Method for Efficient Approximations in Large-Scale Machine Learning
• Query Twice: Dual Mixture Attention Meta Learning for Video Summarization
• Deep Submodular Functions
• Graph Tikhonov Regularization and Interpolation via Random Spanning Forests
• Video Skimming: Taxonomy and Comprehensive Survey
• A Separation Logic for Negative Dependence
• Determinantal Point Process Priors for Bayesian Variable Selection in Linear Regression
• Exact Sampling of Determinantal Point Processes without Eigendecomposition
• Evolutionary approaches to explainable machine learning
• PcDGAN: A Continuous Conditional Diverse Generative Adversarial Network For Inverse Design
• Approximate Inference in Continuous Determinantal Point Processes
• High-performance sampling of generic Determinantal Point Processes
• Accelerated Sampling on Discrete Spaces with Non-Reversible Markov Processes
• Universality in the number variance and counting statistics of the real and symplectic Ginibre ensemble
• Invariant Coupling of Determinantal Measures on Sofic Groups
• Determinantal Point Processes for Mini-Batch Diversification
• Fast Mixing for Discrete Point Processes
• Spatial networks with wireless applications
• Monte Carlo Markov Chain Algorithms for Sampling Strongly Rayleigh Distributions and Determinantal Point Processes
• Optimal DR-Submodular Maximization and Applications to Provable Mean Field Inference
• One Sample Stochastic Frank-Wolfe
• Low-Rank Factorization of Determinantal Point Processes for Recommendation
• Autoregressive neural Slater-Jastrow ansatz for variational Monte Carlo simulation
• Deep Generative Model-based Synthesis of Four-bar Linkage Mechanisms with Target Conditions
• Diversifying Sparsity Using Variational Determinantal Point Processes
• Feature-based Learning for Diverse and Privacy-Preserving Counterfactual Explanations
• Pareto Front-Diverse Batch Multi-Objective Bayesian Optimization
• Diversifying Multi-aspect Search Results Using Simpson's Diversity Index
• The non-stabilizerness of fermionic Gaussian states
• Privacy-preserving Continual Federated Clustering via Adaptive Resonance Theory
• Diversity-Promoting Bayesian Learning of Latent Variable Models
• Tagging like Humans: Diverse and Distinct Image Annotation
• Active Mini-Batch Sampling using Repulsive Point Processes
• A Parallel Double Greedy Algorithm for Submodular Maximization
• Learning compositional models of robot skills for task and motion planning
• Improving robustness and calibration in ensembles with diversity regularization
• Inference for determinantal point processes without spectral knowledge
• Learning Guided Electron Microscopy with Active Acquisition
• Reverse iterative volume sampling for linear regression
• Transfer Learning with Active Sampling for Rapid Training and Calibration in BCI-P300 Across Health States and Multi-centre Data
• Improving End-to-End Sequential Recommendations with Intent-aware Diversification
• Rates of estimation for determinantal point processes
• Maximum likelihood estimation of determinantal point processes
• Representation Online Matters: Practical End-to-End Diversification in Search and Recommender Systems
• On the Parameterized Intractability of Determinant Maximization
• Active model learning and diverse action sampling for task and motion planning
• Orthogonality-Promoting Distance Metric Learning: Convex Relaxation and Theoretical Analysis
• Infinite Probabilistic Databases
• Reconfiguration Problems on Submodular Functions
• A Unified Framework for Generic, Query-Focused, Privacy Preserving and Update Summarization using Submodular Information Measures
• How Good is a Video Summary? A New Benchmarking Dataset and Evaluation Framework Towards Realistic Video Summarization
• Exact Sampling from Determinantal Point Processes
• Sensor Array Design Through Submodular Optimization
• Notes on using Determinantal Point Processes for Clustering with Applications to Text Clustering
• Fast Mixing Markov Chains for Strongly Rayleigh Measures, DPPs, and Constrained Sampling
• Weakly Submodular Maximization Beyond Cardinality Constraints: Does Randomization Help Greedy?
• BOSH: Bayesian Optimization by Sampling Hierarchically
• A Polynomial Time MCMC Method for Sampling from Continuous DPPs
• Viewpoint-aware Video Summarization
• Temporal Tessellation: A Unified Approach for Video Analysis
• Approximately Optimal Spatial Design: How Good is it?
• Discovery of Latent Factors in High-dimensional Data Using Tensor Methods
• Discovering Valuable Items from Massive Data
• FormulaZero: Distributionally Robust Online Adaptation via Offline Population Synthesis
• Some Inapproximability Results of MAP Inference and Exponentiated Determinantal Point Processes
• Distributed Submodular Maximization
• Unsupervised Summarization by Jointly Extracting Sentences and Keywords
• Composable Core-sets for Determinant Maximization: A Simple Near-Optimal Algorithm
• Structured Monte Carlo Sampling for Nonisotropic Distributions via Determinantal Point Processes
• Normal approximation for associated point processes via Stein's method with applications to determinantal point processes
• Spanning Tree Constrained Determinantal Point Processes are Hard to (Approximately) Evaluate
• Exploiting local and global performance of candidate systems for aggregation of summarization techniques
• Optimal Design Emulators: A Point Process Approach
• VBALD - Variational Bayesian Approximation of Log Determinants
• Weighted Point Configurations with Hyperuniformity: An Ecological Example and Models
• Approximation of Functions: Optimal Sampling and Complexity
• Faster Greedy MAP Inference for Determinantal Point Processes
• Learning Diverse Representations for Fast Adaptation to Distribution Shift
• On determinantal point processes with nonsymmetric kernels
• Give me a hint! Navigating Image Databases using Human-in-the-loop Feedback
• Fast Sampling for Strongly Rayleigh Measures with Application to Determinantal Point Processes
• Block-Wise MAP Inference for Determinantal Point Processes with Application to Change-Point Detection
• A Critical Reexamination of Intra-List Distance and Dispersion
• Circulant $L$-ensembles in the thermodynamic limit
• Coverage probability in wireless networks with determinantal scheduling
• Deep Submodular Networks for Extractive Data Summarization
• Analyzing high-dimensional time-series data using kernel transfer operator eigenfunctions
• Fermion Sampling Made More Efficient
• Discrete Sampling using Semigradient-based Product Mixtures
• Mean and variance of the cardinality of particles in polyanalytic Ginibre processes via a quantization method
• Active Preference Learning for Personalized Portfolio Construction
• On distances among Slater Determinant States and Determinantal Point Processes
• Nonparametric estimation of continuous DPPs with kernel methods
• On sampling determinantal and Pfaffian point processes on a quantum computer
• Gibbs Sampling using Anti-correlation Gaussian Data Augmentation, with Applications to L1-ball-type Models
• The Bayesian Low-Rank Determinantal Point Process Mixture Model
• Statistical physics of principal minors: Cavity approach
• Multi-type branching processes with immigration generated by point processes
• Multi-Task Determinantal Point Processes for Recommendation
• Rapid Posterior Exploration in Bayesian Non-negative Matrix Factorization