Deep Metric Learning

[ICCV 2021] Focal Frequency Loss for Image Reconstruction and Synthesis

Angular penalty loss functions in Pytorch (ArcFace, SphereFace, Additive Margin, CosFace)

PyTorch implementation of Soft-DTW: a Differentiable Loss Function for Time-Series in CUDA

A better pytorch-based implementation for the mean structural similarity. Differentiable simpler SSIM and MS-SSIM.

A dependency free library of standardized optimization test functions written in pure Python.

[CVPR 2024] Adaptive Multi-Modal Cross-Entropy Loss for Stereo Matching

Seach Losses of our paper 'Loss Function Discovery for Object Detection via Convergence-Simulation Driven Search', accepted by ICLR 2021.

Co-VeGAN: Complex-Valued Generative Adversarial Network for Compressive Sensing MR Image Reconstruction

Angular triplet center loss implementation in Pytorch.

Tensorflow Implementation of Focal Frequency Loss for Image Reconstruction and Synthesis [ICCV 2021]

Directional Distance Field for Modeling the Difference between 3D Point Clouds

A simple 3-layer fully connected network performing the density ratio estimation using the loss for log-likelihood ratio estimation (LLLR).

Official PyTorch implementation for "PROPEL: Probabilistic Parametric Regression Loss for Convolutional Neural Networks"

Supervised Sliding Window Smoothing Loss Function Based on MS-TCN for Video Segmentation

Tensor-Network Machine Learning with Matrix Product States, trained via a surrogate (projective) loss instead of standard negative log-likelihood

An HR predictive analytics tool for forecasting the likely range of a worker’s future job performance using multiple ANNs with custom loss functions.

[TPAMI 2023] PyTorch code for Sparse Label Smoothing Regularization presented in "Learning Symbolic Model-Agnostic Loss Functions via Meta-Learning". Paper Link: https://arxiv.org/abs/2209.08907

Inverse Supervised Learning

Alternative loss function of binary cross entropy and focal loss