Skip to content
AI.info

AI.info

Advanced techniques

Ensembles, self-supervision, mixtures of experts — with a decision map for when the simple thing has stopped working.

  1. Advanced Techniques: A Decision Map
  2. Ensemble Diversity and Error Correlation
  3. Bagging, Randomization, and Out-of-Bag Evaluation
  4. Boosting: From Weak Learners to Gradient Models
  5. Stacking, Blending, and Weight Averaging
  6. Transfer Learning and Feature Reuse
  7. Domain Adaptation and Domain Generalization
  8. Fine-Tuning and Parameter-Efficient Adaptation
  9. Few-Shot and In-Context Learning
  10. Semi-Supervised Learning: Pseudo-Labels, Consistency, and Confirmation Bias
  11. Self-Supervised Learning: Designing the Objective
  12. Contrastive and Bootstrap Representation Learning
  13. Masked Modeling and Predictive Pretraining
  14. Knowledge Distillation and Teacher–Student Learning
  15. Meta-Learning and Fast Adaptation
  16. Continual Learning and Catastrophic Forgetting
  17. Curriculum Learning, Hard Examples, and Data Scheduling
  18. Actor–Critic Methods and Proximal Policy Optimization
  19. Offline Reinforcement Learning and Imitation Learning
  20. Graph Neural Networks and Message Passing
  21. Graph Tasks, Sampling, and Failure Modes
  22. Multi-Task Learning and Mixture-of-Experts
  23. Evaluating and Debugging Advanced Techniques
  24. Advanced Techniques Capstone