LLMs as Agents
Subscribers:
22,300
Published on ● Video Link: https://www.youtube.com/watch?v=nHI92zDQIxo
0Unlock the next frontier in Reinforcement Learning by turning Large Language Models (LLMs) and Vision‑Language Models (VLMs) into agents themselves! In this deep‑dive, we explore how parametric methods use fine‑tuning techniques like LoRA, adapter weights, and prefix tuning to teach your LLM/VLM from new trajectories, and how non‑parametric methods leverage retrieval‑augmented prompts to guide action choices without modifying model weights.
If you found this helpful, be sure to subscribe for more AI tutorials, give the video a thumbs‑up, and leave your questions or experiences in the comments below.
#AI #MachineLearning #DeepLearning #ReinforcementLearning #LLM #VLM #FineTuning #PromptEngineering
Other Videos By LLMs Explained - Aggregate Intellect - AI.SCIENCE
| 2025-05-19 | Use Cases of State Machines |
| 2025-05-17 | Why Do We Need Sherpa |
| 2025-05-16 | When Should We Use Sherpa? |
| 2025-05-15 | How Do State Machines Work? |
| 2025-05-10 | Best Practices for Prompt Safety |
| 2025-05-09 | What is Data Privacy |
| 2025-05-08 | Best Practices for Protecting Data |
| 2025-05-01 | Strengths, Challenges, and Problem Formulation in RL |
| 2025-04-30 | How LLMs Can Help RL Agents Learn |
| 2025-04-29 | LLM VLM Based Reward Models |
| 2025-04-28 | LLMs as Agents |
| 2025-04-10 | Data Stores, Prompt Repositories, and Memory Management |
| 2025-04-10 | Dynamic Prompting and Retrieval Techniques |
| 2025-04-09 | How to Fine Tune Agents |
| 2025-04-08 | What are Agents |
| 2025-04-02 | Leveraging LLMs for Causal Reasoning |
| 2025-04-01 | Examples of Causal Representation in Computer vision |
| 2025-03-31 | Relationship between Reasoning and Causality |
| 2025-03-30 | Causal Representation Learning |
| 2025-03-18 | Deduplication in DeepSeek R1 |
| 2025-03-17 | What Makes DeepSeek R1 Multi-token Prediction Unique? |