SELFIE
by kaist-dmlab
AI summary
Data refinement method
A method to enhance robustness in deep learning by selectively refining noisy training data and combining it with clean samples.
- stars
- 50
- forks
- 9
- watching
- 7
Similar projects
Found by comparing what the projects do, not just their names.
Model robustness testing
Evaluates and benchmarks the robustness of deep learning models to various corruptions and perturbations in computer vision tasks.
Memory optimizer for deep learning models
A lightweight wrapper around PyTorch to prevent CUDA out-of-memory errors and optimize model execution
Deep learning library
A collection of GPU-accelerated deep learning algorithms implemented in Python
Memory optimizer
A JAX transform that simplifies the training of large language models by reducing memory usage through low-rank adaptation.
Deep Learning Model
A software framework for building deep metric learning models using lifted structured feature embedding
Deep RL Algorithm
An algorithm for deep reinforcement learning that combines model-free and model-based approaches to learn robust value functions.
Point cloud segmentation
A deep learning framework for efficient semantic segmentation of large-scale 3D point clouds
Portfolio optimizer
An algorithm that optimizes portfolio allocation using Reinforcement Learning and Supervised learning.
Segmentation model
A deep learning implementation of an object segmentation algorithm.
Inference optimizer
A deep learning method for optimizing convolutional neural networks by reducing computational cost while improving regularization and inference efficiency.
Object Deblurring
A deep learning framework for deblurring and recovering the shape of fast-moving objects from blurred images
Video stabilizer
This project provides an implementation of video stabilization with iterative optimization using deep learning techniques.
Image harmonizer
Reimplements a deep learning model to harmonize images from different illumination conditions
Noisy label solver
An implementation of a contrastive learning approach to address noisy labels in machine learning models
Model distiller
Automatically trains models from large foundation models to perform specific tasks with minimal human intervention.
