By Π Π°Π²ΠΈΡΠ°Π½Π΄ΠΈΡΠ°Π½ Π‘ΡΠ΄Ρ Π°ΡΡΠ°Π½
Π‘ΡΠ΄Ρ Π°ΡΡΠ°Π½ Π Π°Π²ΠΈΡΠ°Π½Π΄ΠΈΡΠ°Π½, 2020
ΠΠ»ΡΠ±ΠΎΠΊΠΎΠ΅ ΠΎΠ±ΡΡΠ΅Π½ΠΈΠ΅ Ρ ΠΏΠΎΠ΄ΠΊΡΠ΅ΠΏΠ»Π΅Π½ΠΈΠ΅ΠΌ (Reinforcement Learning) ΡΠ²Π»ΡΠ΅ΡΡΡ Π²Π΅Π΄ΡΡΠΈΠΌ Π½Π°ΠΏΡΠ°Π²Π»Π΅Π½ΠΈΠ΅ΠΌ Π² ΠΎΠ±Π»Π°ΡΡΠΈ ΠΈΡΠΊΡΡΡΡΠ²Π΅Π½Π½ΠΎΠ³ΠΎ ΠΈΠ½ΡΠ΅Π»Π»Π΅ΠΊΡΠ°. ΠΡΠΎ ΠΏΡΠ°ΠΊΡΠΈΡΠ΅ΡΠΊΠΎΠ΅ ΡΡΠΊΠΎΠ²ΠΎΠ΄ΡΡΠ²ΠΎ Π½Π° Python ΠΏΠΎΠ·Π²ΠΎΠ»ΠΈΡ ΠΎΡΠ²ΠΎΠΈΡΡ ΠΊΠ°ΠΊ Π±Π°Π·ΠΎΠ²ΡΠ΅, ΡΠ°ΠΊ ΠΈ ΠΏΠ΅ΡΠ΅Π΄ΠΎΠ²ΡΠ΅ Π°Π»Π³ΠΎΡΠΈΡΠΌΡ RL. ΠΡ Π½Π°ΡΠ½Π΅ΡΠ΅ Ρ ΠΎΡΠ½ΠΎΠ², Π²ΠΊΠ»ΡΡΠ°Ρ OpenAI Gym ΠΈ TensorFlow, ΠΈΠ·ΡΡΠΈΡΠ΅ ΠΌΠ°ΡΠΊΠΎΠ²ΡΠΊΠΈΠ΅ ΡΠ΅ΠΏΠΈ, ΠΌΠ΅ΡΠΎΠ΄Ρ ΠΠΎΠ½ΡΠ΅-ΠΠ°ΡΠ»ΠΎ ΠΈ Π΄ΠΈΠ½Π°ΠΌΠΈΡΠ΅ΡΠΊΠΎΠ΅ ΠΏΡΠΎΠ³ΡΠ°ΠΌΠΌΠΈΡΠΎΠ²Π°Π½ΠΈΠ΅. ΠΠ½ΠΈΠ³Π° ΠΏΠΎΠΌΠΎΠΆΠ΅Ρ ΡΠ°Π·ΠΎΠ±ΡΠ°ΡΡΡΡ Π² Π°Π±Π±ΡΠ΅Π²ΠΈΠ°ΡΡΡΠ°Ρ DQN, DRQN, A3C, PPO ΠΈ TRPO, Π° ΡΠ°ΠΊΠΆΠ΅ Π² Π°Π³Π΅Π½ΡΠ°Ρ , ΠΎΠ±ΡΡΠ°ΡΡΠΈΡ ΡΡ Π½Π° ΠΏΡΠ΅Π΄ΠΏΠΎΡΡΠ΅Π½ΠΈΡΡ ΡΠ΅Π»ΠΎΠ²Π΅ΠΊΠ°, DQfD ΠΈ HER.
Sudharsan Ravichandiran, 2020
Deep Reinforcement Learning (RL) is a leading area in artificial intelligence. This practical Python guide covers fundamental and advanced RL algorithms. You will start with the basics, including OpenAI Gym and TensorFlow, and explore Markov chains, Monte Carlo methods, and dynamic programming. The book demystifies acronyms like DQN, DRQN, A3C, PPO, and TRPO, alongside agents learning from human preferences, DQfD, and HER.