MLPapers

MLPapers is a collection of regularly updated reviews of machine learning research. Each topic connects methods, papers and implementations, summarizing how methods work, their assumptions, and what later findings extend or challenge.

Topics

Modeling and inference

• Bayesian inference
• Bayesian networks
• Causal inference
• Ensemble learning
• Generative models
• Modeling
• Symbolic regression

Networks and representations

• Feature extraction
• Geometric deep learning
• Graph neural networks
• Multiview learning
• Neural networks
• Neurosymbolic learning

Data analysis

• Anomaly detection
• Clustering
• Feature selection
• Missing data

Training and adaptation

• AutoML
• Continual learning
• Meta-learning
• Optimization
• Transfer learning

Evaluation and reliability

• Evaluation
• Generalization
• Interpretability

Distributed learning and privacy

• Distributed learning
• Federated learning
• Machine unlearning
• Privacy-preserving learning

Forecasting and control

• Model predictive control
• Reinforcement learning
• Time series forecasting

Domains and applications

• Computer vision
• Natural language processing
• Scientific machine learning