Data to 7 Oct 2026

Reparameterizable Subset Sampling via Continuous Relaxations

arXiv 1901.10517 ↗First posted 29 Jan 2019cs.LG stat.MLcited by 69

Not on the map. Why?

Abstract (arXiv)

Many machine learning tasks require sampling a subset of items from a collection based on a parameterized distribution. The Gumbel-softmax trick can be used to sample a single item, and allows for low-variance reparameterized gradients with respect to the parameters of the underlying distribution. However, stochastic optimization involving subset sampling is typically not reparameterizable. To overcome this limitation, we define a continuous relaxation of subset sampling that provides reparameterization gradients by generalizing the Gumbel-max trick. We use this approach to sample subsets of features in an instance-wise feature selection task for model interpretability, subsets of neighbors to implement a deep stochastic k-nearest neighbors model, and sub-sequences of neighbors to implement parametric t-SNE by directly comparing the identities of local neighbors. We improve performance in all these tasks by incorporating subset sampling in end-to-end training.

Key facts. Cited by 69 papers in the map (data through 7 Oct 2026). First posted on arXiv 29 Jan 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 10; 2022: 12; 2023: 16; 2024: 8; 2025: 12; 2026: 11.

69 papers.

Sort:

By area:

2023Data Selection for Language Models via Importance Resamplingcited by 322
2021DSelect-k: Differentiable Selection in the Mixture of Experts with Applications to Multi-Task Learningcited by 114
2021Implicit MLE: Backpropagating Through Discrete Exponential Family Distributionscited by 97
2021A Review of the Gumbel-max Trick and its Extensions for Discrete Stochasticity in Machine Learningcited by 82
2021Differentiable Patch Selection for Image Recognitioncited by 68
2023Diffusion Models With Learned Adaptive Noisecited by 50
2021Differentiable Subset Pruning of Transformer Headscited by 46
2024Adaptive Data Optimization: Dynamic Sample Selection with Scaling Lawscited by 46
2022SIMPLE: A Gradient Estimator for k-Subset Samplingcited by 34
2022Backpropagation through Combinatorial Algorithms: Identity with Projection Workscited by 29
2022AdaProp: Learning Adaptive Propagation for Graph Neural Network based Knowledge Graph Reasoningcited by 28
2023Probabilistically Rewired Message-Passing Neural Networkscited by 26

38 more

2021DSKReG: Differentiable Sampling on Knowledge Graph for Recommendation with Relational GNNcited by 19
2024End-to-End Video Question Answering with Frame Scoring Mechanisms and Adaptive Samplingcited by 19
2025A Comprehensive Survey on Self-Interpretable Neural Networkscited by 18
2023On Strengthening and Defending Graph Reconstruction Attack with Markov Chain Approximationcited by 16
2023Differentiable Multi-Target Causal Bayesian Experimental Designcited by 15
2024Probabilistic Graph Rewiring via Virtual Nodescited by 15
2025Sparse Logit Sampling: Accelerating Knowledge Distillation in LLMscited by 14
2022Adaptive Perturbation-Based Gradient Estimation for Discrete Latent Variable Modelscited by 12
2022Calibration of Machine Reading Systems at Scalecited by 11
2025Relational Conformal Prediction for Correlated Time Seriescited by 11
2022Stochastic smoothing of the top-K calibrated hinge loss for deep imbalanced classificationcited by 10
2025LLaVA-ReID: Selective Multi-image Questioner for Interactive Person Re-Identificationcited by 9
2021Adversarial Contrastive Pre-training for Protein Sequencescited by 8
2021Leveraging Recursive Gumbel-Max Trick for Approximate Inference in Combinatorial Spacescited by 7
2024Dynamic Encoder Size Based on Data-Driven Layer-wise Pruning for Speech Recognitioncited by 7
2022Differentiable Mathematical Programming for Object-Centric Representation Learningcited by 6
2021Dynamic Probabilistic Pruning: A general framework for hardware-constrained pruning at different granularitiescited by 5
2022Learning Group Importance using the Differentiable Hypergeometric Distributioncited by 5
2023Node Injection for Class-specific Network Poisoningcited by 5
2025LookWhere? Efficient Visual Recognition by Learning Where to Look and What to See from Self-Supervisioncited by 5
2022Accelerated Intravascular Ultrasound Imaging using Deep Reinforcement Learningcited by 4
2023Token Sparsification for Faster Medical Image Segmentationcited by 4
2023Rumor Detection with Diverse Counterfactual Evidencecited by 4
2025Long-VMNet: Accelerating Long-Form Video Understanding via Fixed Memorycited by 4
2021Joint optimization of system design and reconstruction in MIMO radar imagingcited by 3
2023Differentiable Random Partition Modelscited by 3
2023Decongestion by Representation: Learning to Improve Economic Welfare in Marketplacescited by 3
2024Unveiling Latent Causal Rules: A Temporal Point Process Approach for Abnormal Event Explanationcited by 3
2022Weakly-Supervised End-to-End CAD Retrieval to Scan Objectscited by 2
2023Copula for Instance-wise Feature Selection and Rankingcited by 2
2023Everything Perturbed All at Once: Enabling Differentiable Graph Attackscited by 2
2023All Things Considered: Detecting Partisan Events from News Media with Cross-Article Comparisoncited by 2
2025Gumbel Reranking: Differentiable End-to-End Reranker Optimizationcited by 2
2022MarkerMap: nonlinear marker selection for single-cell studiescited by 1
2022Explainability as statistical inferencecited by 1
2023Language as a Latent Sequence: deep latent variable models for semi-supervised paraphrase generationcited by 1
2024Revisiting Score Function Estimators for k-Subset Samplingcited by 1
2024Enhancing Topic Interpretability for Neural Topic Modeling through Topic-wise Contrastive Learningcited by 1

50 most cited of 69 All 69 (JSON)