Reinforcement Learning Avanzado: de DQN a SAC
Learn to build intelligent agents using Deep Reinforcement Learning (DRL) with PyTorch. Explore DDPG, TD3, SAC, NAF & HER algorithms.
Course Content
Reinforcement Learning Avanzado: de DQN a SAC is an all-levels Udemy course taught by Escape Velocity Labs. It runs 7h 54m, is taught in Spanish, and includes a certificate of completion. Learners rate it 4.9 out of 5 from 10 reviews, with 111 students enrolled. It is a paid course, last updated May 2025.
Unlock the power of Advanced Reinforcement Learning with this comprehensive Udemy course! Dive deep into implementing cutting-edge Deep Reinforcement Learning algorithms using Python, PyTorch, and PyTorch Lightning.
This course takes you from the ground up, teaching you to build adaptive agents that master control tasks through experience-based learning. You'll combine powerful Neural Networks and Deep Learning techniques to create intelligent systems capable of sophisticated decision-making.
Get introduced to the latest advancements in Reinforcement Learning and prepare for advanced topics in future courses. We focus on practical skills – you'll implement algorithms from scratch in Jupyter Notebooks after mastering the core concepts.
Course Highlights:
Master Core Concepts: Solidify your understanding with thorough reviews of fundamental topics, including:
- Markov Decision Processes (MDPs): Understand the foundation of RL.
- Q-Learning: A classic RL algorithm.
- Introduction to Neural Networks: Build a strong base in deep learning.
- Deep Q-Learning: Combine Q-Learning with deep neural networks.
- Policy Gradient Methods: Learn how to directly optimize policies.
Deep Reinforcement Learning Expertise: Explore state-of-the-art algorithms:
- PyTorch Lightning: Streamline your deep learning workflows.
- Hyperparameter Tuning with Optuna: Optimize your models for peak performance.
- Deep Q-Learning for Continuous Action Spaces (NAF): Tackle complex control problems.
- Deep Deterministic Policy Gradient (DDPG): Learn to optimize deterministic policies.
- Twin Delayed DDPG (TD3): Improve stability and performance with TD3.
- Soft Actor-Critic (SAC): Explore maximum entropy reinforcement learning.
- Hindsight Experience Replay (HER): Enhance sample efficiency with HER.
This course is perfect for anyone looking to develop practical skills in Reinforcement Learning, build intelligent agents, and advance their knowledge in AI and Machine Learning.
Want AI in your product? Hire the team behind TutorialSearch.io
LLM features, RAG, agents and automation wired into a real product — built by the devs who shipped this site.