Methods › General › Loss Functions › Adaptive Loss
Adaptive Robust Loss
Adaptive Loss
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
The Robust Loss is a generalization of the Cauchy/Lorentzian, Geman-McClure, Welsch/Leclerc, generalized Charbonnier, Charbonnier/pseudo-Huber/L1-L2, and L2 loss functions. By introducing robustness as a continuous parameter, the loss function allows algorithms built around robust loss minimization to be generalized, which improves performance on basic vision tasks such as registration and clustering. Interpreting the loss as the negative log of a univariate density yields a general probability distribution that includes normal and Cauchy distributions as special cases. This probabilistic interpretation enables the training of neural networks in which the robustness of the loss automatically adapts itself during training, which improves performance on learning-based tasks such as generative image synthesis and unsupervised monocular depth estimation, without requiring any manual parameter tuning.
Papers archive 2025-07-28
30 shown of 73, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.
-
Loss-Guided Model Sharing and Local Learning Correction in Decentralized Federated Learning for Crop Disease Classification 29 May 2025 · 0 repositories · arXiv:2505.23063
-
CALF: A Conditionally Adaptive Loss Function to Mitigate Class-Imbalanced Segmentation 6 Apr 2025 · 0 repositories · arXiv:2504.04458
-
$μ$KE: Matryoshka Unstructured Knowledge Editing of Large Language Models 1 Apr 2025 · 0 repositories · arXiv:2504.01196
-
MagicInfinite: Generating Infinite Talking Videos with Your Words and Voice 7 Mar 2025 · 0 repositories · arXiv:2503.05978
-
Variance-Aware Loss Scheduling for Multimodal Alignment in Low-Data Settings 5 Mar 2025 · 0 repositories · arXiv:2503.03202
-
APT: Adaptive Personalized Training for Diffusion Models with Limited Data 1 Jan 2025 · 0 repositories
-
Never Reset Again: A Mathematical Framework for Continual Inference in Recurrent Neural Networks 20 Dec 2024 · 0 repositories · arXiv:2412.15983
-
Compositional Zero-Shot Learning with Contextualized Cues and Adaptive Contrastive Training 10 Dec 2024 · 0 repositories · arXiv:2412.07161
-
FedDUAL: A Dual-Strategy with Adaptive Loss and Dynamic Aggregation for Mitigating Data Heterogeneity in Federated Learning 5 Dec 2024 · 1 repository · arXiv:2412.04416
-
Multi-scale Cascaded Large-Model for Whole-body ROI Segmentation 23 Nov 2024 · 1 repository · arXiv:2411.15526
-
A mixed gas concentration regression prediction method based on RESHA-ALW 1 Nov 2024 · 0 repositories
-
Network scaling and scale-driven loss balancing for intelligent poroelastography 27 Oct 2024 · 0 repositories · arXiv:2411.08886
-
Spatial-Temporal Mixture-of-Graph-Experts for Multi-Type Crime Prediction 24 Sep 2024 · 0 repositories · arXiv:2409.15764
-
A Physics Informed Neural Network (PINN) Methodology for Coupled Moving Boundary PDEs 17 Sep 2024 · 1 repository · arXiv:2409.10910
-
Real-time Accident Anticipation for Autonomous Driving Through Monocular Depth-Enhanced 3D Modeling 2 Sep 2024 · 0 repositories · arXiv:2409.01256
-
Trust And Balance: Few Trusted Samples Pseudo-Labeling and Temperature Scaled Loss for Effective Source-Free Unsupervised Domain Adaptation 1 Sep 2024 · 0 repositories · arXiv:2409.00741
-
Hybrid Classification-Regression Adaptive Loss for Dense Object Detection 30 Aug 2024 · 0 repositories · arXiv:2408.17182
-
ALTBI: Constructing Improved Outlier Detection Models via Optimization of Inlier-Memorization Effect 19 Aug 2024 · 0 repositories · arXiv:2408.09791
-
Coordinate-Aware Thermal Infrared Tracking Via Natural Language Modeling 11 Jul 2024 · 1 repository · arXiv:2407.08265
-
Modeling Temporal Dependencies within the Target for Long-Term Time Series Forecasting 7 Jun 2024 · 0 repositories · arXiv:2406.04777
-
Thyroid ultrasound diagnosis improvement via multi-view self-supervised learning and two-stage pre-training 18 Feb 2024 · 0 repositories · arXiv:2402.11497
-
Enriched Physics-informed Neural Networks for Dynamic Poisson-Nernst-Planck Systems 1 Feb 2024 · 0 repositories · arXiv:2402.01768
-
Weakly Supervised Semantic Segmentation for Driving Scenes 21 Dec 2023 · 1 repository · arXiv:2312.13646
-
UMedNeRF: Uncertainty-aware Single View Volumetric Rendering for Medical Neural Radiance Fields 10 Nov 2023 · 0 repositories · arXiv:2311.05836
-
Multi-task Learning for Optical Coherence Tomography Angiography (OCTA) Vessel Segmentation 3 Nov 2023 · 0 repositories · arXiv:2311.02266
-
L2T-DLN: Learning to Teach with Dynamic Loss Network 30 Oct 2023 · 0 repositories · arXiv:2310.19313
-
Combating Label Noise With A General Surrogate Model For Sample Selection 16 Oct 2023 · 0 repositories · arXiv:2310.10463
-
Enhanced LFTSformer: A Novel Long-Term Financial Time Series Prediction Model Using Advanced Feature Engineering and the DS Encoder Informer Architecture 3 Oct 2023 · 0 repositories · arXiv:2310.01884
-
FourierLoss: Shape-Aware Loss Function with Fourier Descriptors 21 Sep 2023 · 0 repositories · arXiv:2309.12106
-
Weak-PDE-LEARN: A Weak Form Based Approach to Discovering PDEs From Noisy, Limited Data 9 Sep 2023 · 1 repository · arXiv:2309.04699
Tasks archive 2025-07-28
20 shown of 104 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections