The rest of the modeling landscape a ML engineer should recognize — generative models, recommenders, and the fundamentally different paradigm of learning from reward. Part of the ML Engineer track.

  1. Autoencoders & GANs — [Advanced] — learning compressed representations, and generative adversarial networks
  2. Recommendation Systems — [Intermediate] — collaborative filtering and embeddings for “users who liked X”
  3. Reinforcement Learning — [Advanced] — learning by acting and receiving reward: Q-learning, policy gradients, and where RL fits
↑ Back to 02 — ML Engineer