mlsanityData to 13 Oct 2026

RLGS: Reinforcement Learning-Based Adaptive Hyperparameter Tuning for Gaussian Splatting

arXiv 2508.04078 ↗First posted 6 Aug 2025cs.GR cs.CV cs.LGcited by 3

On the map: Deep reinforcement learning methods → RL for robot locomotion and manipulation

Abstract (arXiv)

Hyperparameter tuning in 3D Gaussian Splatting (3DGS) is a labor-intensive and expert-driven process, often resulting in inconsistent reconstructions and suboptimal results. We propose RLGS, a plug-and-play reinforcement learning framework for adaptive hyperparameter tuning in 3DGS through lightweight policy modules, dynamically adjusting critical hyperparameters such as learning rates and densification thresholds. The framework is model-agnostic and seamlessly integrates into existing 3DGS pipelines without architectural modifications. We demonstrate its generalization ability across multiple state-of-the-art 3DGS variants, including Taming-3DGS and 3DGS-MCMC, and validate its robustness across diverse datasets. RLGS consistently enhances rendering quality. For example, it improves Taming-3DGS by 0.7dB PSNR on the Tanks and Temple (TNT) dataset, under a fixed Gaussian budget, and continues to yield gains even when baseline performance saturates. Our results suggest that RLGS provides an effective and general solution for automating hyperparameter tuning in 3DGS training, bridging a gap in applying reinforcement learning to 3DGS.

Key facts. Cited by 3 papers in the map (data through 13 Oct 2026). Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods. Builds on 29 of its 51 references that are in the map. First posted on arXiv 6 Aug 2025.

Builds on

29 of 51 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Query-based Semantic Gaussian Field for Scene Representation in Reinforcement Learning4 Jun
  2. 2024Reinforcement Learning with Generalizable Gaussian Splatting18 Mar

Other references · 27

  1. 2017Proximal Policy Optimization Algorithmscited by 14,592
  2. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,082
  3. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,170
  4. 2016Neural Architecture Search with Reinforcement Learningcited by 1,496
  5. 2014On the Properties of Neural Machine Translation: Encoder-Decoder Approachescited by 1,303
  6. 20242D Gaussian Splatting for Geometrically Accurate Radiance Fieldscited by 1,070
  7. 2019Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representationscited by 864
  8. 2018Efficient Neural Architecture Search via Parameter Sharingcited by 834
  9. 2016Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimizationcited by 693
  10. 2016Learning to learn by gradient descent by gradient descentcited by 617
  11. 2023DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Visioncited by 521
  12. 2023LightGaussian: Unbounded 3D Gaussian Compression with 15x Reduction and 200+ FPScited by 426
15 more
  1. 2018DeepVoxels: Learning Persistent 3D Feature Embeddingscited by 403
  2. 2019SynSin: End-to-end View Synthesis from a Single Imagecited by 364
  3. 2017Population Based Training of Neural Networkscited by 359
  4. 2019DeepView: View Synthesis with Learned Gradient Descentcited by 253
  5. 2017Automated Curriculum Learning for Neural Networkscited by 252
  6. 2024VastGaussian: Vast 3D Gaussians for Large Scene Reconstructioncited by 209
  7. 2024Taming 3DGS: High-Quality Radiance Fields with Limited Resourcescited by 201
  8. 2018Meta-Gradient Reinforcement Learningcited by 158
  9. 2024EVER: Exact Volumetric Ellipsoid Rendering for Real-time View Synthesiscited by 47
  10. 20243DGS-LM: Faster Gaussian-Splatting Optimization with Levenberg-Marquardtcited by 43
  11. 2018Variance Reduction for Reinforcement Learning in Input-Driven Environmentscited by 30
  12. 2024Sort-free Gaussian Splatting via Weighted Sum Renderingcited by 29
  13. 2024RL-GSBridge: 3D Gaussian Splatting Based Real2Sim2Real Method for Robotic Manipulation Learningcited by 21
  14. 2025Does 3D Gaussian Splatting Need Accurate Volumetric Rendering?cited by 19
  15. 2025Matrix-free Second-order Optimization of Gaussian Splats with Residual Samplingcited by 5

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025A-TDOM: Active TDOM via On-the-Fly 3DGScited by 1
  2. 2026TreeDGS: Aerial Gaussian Splatting for Distant DBH Measurementcited by 0
  3. 2026Gauss What You Need: Compact Gaussian Splatting Across Scene Scalescited by 0