Methods › General › Loss Functions › Huber loss
Huber loss
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
The Huber loss function describes the penalty incurred by an estimation procedure f. Huber (1964) defines the loss function piecewise by[1]
L δ ( a ) = { 1 2 a 2 for | a | ≤ δ , δ ⋅ ( | a | − 1 2 δ ) , otherwise. {\displaystyle L_{δ }(a)={\begin{cases}{\frac {1}{2}}{a^{2}}&for |a|≤ δ ,\\\delta · \left(|a|-{\frac {1}{2}}δ \right),&otherwise.\end{cases}}}
This function is quadratic for small values of a, and linear for large values, with equal values and slopes of the different sections at the two points where | a | = δ |a|=δ . The variable a often refers to the residuals, that is to the difference between the observed and predicted values a = y − f ( x ) a=y-f(x), so the former can be expanded to[2]
L δ ( y , f ( x ) ) = { 1 2 ( y − f ( x ) ) 2 for | y − f ( x ) | ≤ δ , δ ⋅ ( | y − f ( x ) | − 1 2 δ ) , otherwise. {\displaystyle L_{δ }(y,f(x))={\begin{cases}{\frac {1}{2}}(y-f(x))^{2}&for |y-f(x)|≤ δ ,\\\delta \ · \left(|y-f(x)|-{\frac {1}{2}}δ \right),&otherwise.\end{cases}}}
The Huber loss is the convolution of the absolute value function with the rectangular function, scaled and translated. Thus it "smoothens out" the former's corner at the origin.
.. math:: ℓ(x, y) = L = {l_1, ..., l_N}^T
with
.. math::
l_n = \begin{cases}
0.5 (x_n - y_n)^2, & if |x_n - y_n| < delta \\
delta * (|x_n - y_n| - 0.5 * delta), & otherwise
\end{cases}
In Syntology name not resolved (unknown_name). The lookup matched no method, paper or venue by this name; that asserts nothing about whether code exists.
Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.
Papers archive 2025-07-28
archive papers tagged: 77 · with a code link: 13 · where Syntology ran a sample: 2 (2 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument) Syntology
Show: all tagged papersonly where code ran (2 of 77 tagged: 2 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument)
Page 1 of 1: papers 1 to 77 of 77, newest first by the archive's date (ties by slug), in archive order.
Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.
-
SHORE: A Long-term User Lifetime Value Prediction Model in Digital Games 12 Jun 2025 · 0 repositories · arXiv:2506.10487
-
Heavy Lasso: sparse penalized regression under heavy-tailed noise via data-augmented soft-thresholding 9 Jun 2025 · 1 repository · arXiv:2506.07790
-
Differentially Private Sparse Linear Regression with Heavy-tailed Responses 7 Jun 2025 · 0 repositories · arXiv:2506.06861
-
A Surrogate Model for the Forward Design of Multi-layered Metasurface-based Radar Absorbing Structures 14 May 2025 · 0 repositories · arXiv:2505.09251
-
A Piecewise Lyapunov Analysis of Sub-quadratic SGD: Applications to Robust and Quantile Regression 11 Apr 2025 · 0 repositories · arXiv:2504.08178
-
Asymptotics of Non-Convex Generalized Linear Models in High-Dimensions: A proof of the replica formula 27 Feb 2025 · 0 repositories · arXiv:2502.20003
-
Explainable AI based System for Supply Air Temperature Forecast 9 Jan 2025 · 0 repositories · arXiv:2501.05163
-
Robust Sparse Regression with Non-Isotropic Designs 31 Oct 2024 · 0 repositories · arXiv:2410.23937
-
Residual-based Adaptive Huber Loss (RAHL) -- Design of an improved Huber loss for CQI prediction in 5G networks 27 Aug 2024 · 0 repositories · arXiv:2408.14718
-
A Huber Loss Minimization Approach to Mean Estimation under User-level Differential Privacy 22 May 2024 · 0 repositories · arXiv:2405.13453
-
Robust deep learning from weakly dependent data 8 May 2024 · 0 repositories · arXiv:2405.05081
-
H-Consistency Guarantees for Regression 28 Mar 2024 · 0 repositories · arXiv:2403.19480
-
A Unified Framework for Center-based Clustering of Distributed Data 2 Feb 2024 · 0 repositories · arXiv:2402.01302
-
A Robust Quantile Huber Loss With Interpretable Parameter Adjustment In Distributional Reinforcement Learning 4 Jan 2024 · 1 repository · arXiv:2401.02325
-
Robust Non-parametric Knowledge-based Diffusion Least Mean Squares over Adaptive Networks 3 Dec 2023 · 0 repositories · arXiv:2312.01299
-
Towards Robust Offline Reinforcement Learning under Diverse Data Corruption 19 Oct 2023 · 2 repositories · arXiv:2310.12955Syntology official (archive's flag): 2 ran · 5 ran (of which 2 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 0 violated, 2 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified (of 9 harvested samples) · 9 pointer-only (licence)
-
Dual Latent State Learning: Exploiting Regional Network Similarities for QoS Prediction 7 Oct 2023 · 0 repositories · arXiv:2310.05988
-
High-dimensional robust regression under heavy-tailed data: Asymptotics and Universality 28 Sep 2023 · 0 repositories · arXiv:2309.16476
-
A Huber Loss Minimization Approach to Byzantine Robust Federated Learning 24 Aug 2023 · 0 repositories · arXiv:2308.12581
-
Machine learning in parameter estimation of nonlinear systems 23 Aug 2023 · 0 repositories · arXiv:2308.12393
-
Multi-Task Pseudo-Label Learning for Non-Intrusive Speech Quality Assessment Model 18 Aug 2023 · 0 repositories · arXiv:2308.09262
-
Investigating the Edge of Stability Phenomenon in Reinforcement Learning 9 Jul 2023 · 0 repositories · arXiv:2307.04210
-
SDR-GAIN: A High Real-Time Occluded Pedestrian Pose Completion Method for Autonomous Driving 6 Jun 2023 · 0 repositories · arXiv:2306.03538
-
Asymptotic Characterisation of Robust Empirical Risk Minimisation Performance in the Presence of Outliers 30 May 2023 · 1 repository · arXiv:2305.18974
-
Distributional Reinforcement Learning with Dual Expectile-Quantile Regression 26 May 2023 · 0 repositories · arXiv:2305.16877
-
Robust Nonparametric Regression under Poisoning Attack 26 May 2023 · 0 repositories · arXiv:2305.16771
-
Computationally Efficient and Statistically Optimal Robust High-Dimensional Linear Regression 10 May 2023 · 0 repositories · arXiv:2305.06199
-
Directly Optimizing IoU for Bounding Box Localization 14 Apr 2023 · 0 repositories · arXiv:2304.07256
-
Robust convex biclustering with a tuning-free method 6 Dec 2022 · 1 repository · arXiv:2212.03122
-
Inflexible Multi-Asset Hedging of incomplete market 2 Nov 2022 · 0 repositories · arXiv:2211.00948
-
DEQGAN: Learning the Loss Function for PINNs with Generative Adversarial Networks 15 Sep 2022 · 1 repository · arXiv:2209.07081
-
Statistical Properties of the log-cosh Loss Function Used in Machine Learning 9 Aug 2022 · 0 repositories · arXiv:2208.04564
-
Sharp Constants in Uniformity Testing via the Huber Statistic 21 Jun 2022 · 0 repositories · arXiv:2206.10722
-
Introducing the Huber mechanism for differentially private low-rank matrix completion 16 Jun 2022 · 0 repositories · arXiv:2206.07910
-
Robust Matrix Completion with Heavy-tailed Noise 9 Jun 2022 · 0 repositories · arXiv:2206.04276
-
Robust Losses for Learning Value Functions 17 May 2022 · 0 repositories · arXiv:2205.08464
-
A Huber loss-based super learner with applications to healthcare expenditures 13 May 2022 · 1 repository · arXiv:2205.06870
-
Re-examining Distillation For Continual Object Detection 4 Apr 2022 · 0 repositories · arXiv:2204.01407
-
How do noise tails impact on deep ReLU networks? 20 Mar 2022 · 0 repositories · arXiv:2203.10418
-
Computationally Efficient and Statistically Optimal Robust Low-rank Matrix and Tensor Estimation 2 Mar 2022 · 0 repositories · arXiv:2203.00953
-
Nonconvex Extension of Generalized Huber Loss for Robust Learning and Pseudo-Mode Statistics 22 Feb 2022 · 0 repositories · arXiv:2202.11141
-
Gradient Based Clustering 1 Feb 2022 · 0 repositories · arXiv:2202.00720
-
Probabilistic Regression with Huber Distributions 19 Nov 2021 · 0 repositories · arXiv:2111.10296
-
Active Linear Regression for ℓₚ Norms and Beyond 9 Nov 2021 · 0 repositories · arXiv:2111.04888
-
Consistent Estimation for PCA and Sparse Regression with Oblivious Outliers 4 Nov 2021 · 0 repositories · arXiv:2111.02966
-
FREGAN : an application of generative adversarial networks in enhancing the frame rate of videos 1 Nov 2021 · 0 repositories · arXiv:2111.01105
-
Robust High-Dimensional Regression with Coefficient Thresholding and its Application to Imaging Data Analysis 30 Sep 2021 · 0 repositories · arXiv:2109.14856
-
Finetuning Transformer Models to Build ASAG System 25 Sep 2021 · 0 repositories · arXiv:2109.12300
-
CFEAR Radarodometry - Conservative Filtering for Efficient and Accurate Radar Odometry 16 Sep 2021 · 2 repositories
-
Generalized Huber Loss for Robust Learning and its Efficient Minimization for a Robust Statistics 28 Aug 2021 · 0 repositories · arXiv:2108.12627
-
Derivatives and residual distribution of regularized M-estimators with application to adaptive tuning 11 Jul 2021 · 0 repositories · arXiv:2107.05143
-
Asymptotic normality of robust M-estimators with convex penalty 8 Jul 2021 · 0 repositories · arXiv:2107.03826
-
Support estimation in high-dimensional heteroscedastic mean regression 3 Nov 2020 · 0 repositories · arXiv:2011.01591
-
Adversarial Robust Low Rank Matrix Estimation: Compressed Sensing and Matrix Completion 25 Oct 2020 · 0 repositories · arXiv:2010.13018
-
Multiple Pedestrians and Vehicles Tracking in Aerial Imagery: A Comprehensive Study 19 Oct 2020 · 0 repositories · arXiv:2010.09689
-
Consistent regression when oblivious outliers overwhelm 30 Sep 2020 · 0 repositories · arXiv:2009.14774
-
Risk Bounds for Robust Deep Learning 14 Sep 2020 · 0 repositories · arXiv:2009.06202
-
Unsupervised Learning of Solutions to Differential Equations with Generative Adversarial Networks 21 Jul 2020 · 1 repository · arXiv:2007.11133
-
Robust estimation with Lasso when outputs are adversarially contaminated 13 Apr 2020 · 0 repositories · arXiv:2004.05990
-
DC-DistADMM: ADMM Algorithm for Constrained Distributed Optimization over Directed Graphs 30 Mar 2020 · 0 repositories · arXiv:2003.13742
-
An Alternative Probabilistic Interpretation of the Huber Loss 5 Nov 2019 · 0 repositories · arXiv:1911.02088
-
Duality in RKHSs with Infinite Dimensional Outputs: Application to Robust Losses 10 Oct 2019 · 0 repositories · arXiv:1910.04621
-
DeepTIO: A Deep Thermal-Inertial Odometry with Visual Hallucination 16 Sep 2019 · 0 repositories · arXiv:1909.07231
-
RobustTrend: A Huber Loss with a Combined First and Second Order Difference Regularization for Time Series Trend Filtering 10 Jun 2019 · 1 repository · arXiv:1906.03751
-
Stochastic Gradients for Large-Scale Tensor Decomposition 4 Jun 2019 · 0 repositories · arXiv:1906.01687
-
Exponentially-Modified Gaussian Mixture Model: Applications in Spectroscopy 14 Feb 2019 · 0 repositories · arXiv:1902.05601
-
Scale calibration for high-dimensional robust regression 6 Nov 2018 · 0 repositories · arXiv:1811.02096
-
Towards a Zero-One Law for Column Subset Selection 4 Nov 2018 · 1 repository · arXiv:1811.01442
-
A Novel Approach to Quantized Matrix Completion Using Huber Loss Measure 29 Oct 2018 · 0 repositories · arXiv:1810.12460
-
Approach for Video Classification with Multi-label on YouTube-8M Dataset 27 Aug 2018 · 0 repositories · arXiv:1808.08671
-
Risk-Averse Classification 30 Apr 2018 · 0 repositories · arXiv:1805.00119
-
Adaptive Regularization of Some Inverse Problems in Image Analysis 9 May 2017 · 0 repositories · arXiv:1705.03350
-
Active Regression with Adaptive Huber Loss 5 Jun 2016 · 0 repositories · arXiv:1606.01568
-
Deeper Depth Prediction with Fully Convolutional Residual Networks 1 Jun 2016 · 18 repositories · arXiv:1606.00373Syntology official (archive's flag): 3 ran · 4 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified (of 5 harvested samples)
-
Video2GIF: Automatic Generation of Animated GIFs from Video 16 May 2016 · 1 repository · arXiv:1605.04850
-
Automatic Inference of the Quantile Parameter 12 Nov 2015 · 0 repositories · arXiv:1511.03990
-
Semismooth Newton Coordinate Descent Algorithm for Elastic-Net Penalized Huber Loss Regression and Quantile Regression 9 Sep 2015 · 0 repositories · arXiv:1509.02957
Tasks archive 2025-07-28
20 shown of 71 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