Discover AI Courses

No AI Self-Improvement with Reinforcement Learning - Limitations of RLVR

Explore the limitations of Reinforcement Learning for AI self-evolution, revealing why RLVR fails to develop new reasoning capabilities beyond a model's initial training.

• YouTube
• 30 minutes
• Self-Paced
• Free Video

AI Will Destroy My Job as a Creative - My Plan to Survive

Discover strategies to adapt and thrive as a creative professional in the face of AI disruption, with insights from recent research on synthetic continued pretraining.

• YouTube
• 32 minutes
• Self-Paced
• Free Video

MCP and A2A Fail - Not for the Reasons You Think

Explore strange phenomena in neural networks, including how new knowledge integrates in transformer architectures and data-efficient learning for AI, with insights on memorization vs. reasoning.

• YouTube
• 40 minutes
• Self-Paced
• Free Video

CODE RED: TTRL Unlocks AI Self-Evolution

Explore the boundaries of AI self-learning through TTRL methodology, examining the limits of self-rewarding and self-referencing reinforcement learning in language models.

• YouTube
• 33 minutes
• Self-Paced
• Free Video

Off-Policy "Zero RL" Explained in Simple Terms

Discover how off-policy reinforcement learning compares to SFT for AI reasoning, exploring the LUFFY approach that integrates on-policy and off-policy zero RL for effective knowledge transfer without traditional RL methods.

• YouTube
• 46 minutes
• Self-Paced
• Free Video

TEXAS: Fine-Tuning Is for Cowards - Do RL

Dive into the debate between Supervised Fine-Tuning and Reinforcement Learning for AI reasoning, exploring research findings on which approach yields better results for vision-language models.

• YouTube
• 33 minutes
• Self-Paced
• Free Video

Multi-LLM Multi-Agents are Cheaper and Better

Explore the benefits of heterogeneous LLMs in multi-agent systems, examining research that suggests diverse language models can improve performance while reducing costs.

• YouTube
• 19 minutes
• Self-Paced
• Free Video

Claude Sonnet 4 Extreme Logic Test

Discover how Claude Sonnet 4 performs on extreme logic tests, evaluating the new Anthropic model's causal reasoning capabilities in this comprehensive assessment.

• YouTube
• 19 minutes
• Self-Paced
• Free Video

Structured Agent Distillation for Large Language Models

Discover how Structured Agent Distillation (SAD) enhances large language model performance before quantization, presented by researchers from MIT, Harvard, CMU, and other institutions.

• YouTube
• 18 minutes
• Self-Paced
• Free Video

Graph Topology for Securing Wall Street's AI Agents and Multi-agent Systems

Explore how graph topology can secure AI agents in financial markets, defend against malicious attacks, and protect multi-agent systems through topology-guided security frameworks.

• YouTube
• 27 minutes
• Self-Paced
• Free Video

Poisoned Agents: AI Safety and Security - Toxic Prompts

Explore AI safety and security through real-world examples of prompt injection attacks on ChatGPT, learning how to protect against these threats and understand the risks involved.

• YouTube
• 20 minutes
• Self-Paced
• Free Video

Security Breach with RAG: Demo with GPT-4.1

Discover how even advanced LLMs like GPT-4.1 struggle with logical inconsistencies in external data inputs during RAG processes, demonstrating security vulnerabilities that structured reasoning templates cannot fully address.

• YouTube
• 11 minutes
• Self-Paced
• Free Video

When Smart AI Models Overthink Stupid Data - AI TRAP

Discover how AI models with advanced reasoning capabilities can overthink ill-posed questions, leading to redundant thinking patterns and inefficient responses—a critical flaw in current LLM training approaches.

• YouTube
• 27 minutes
• Self-Paced
• Free Video

How to Code Agents with Google's NEW ADK

Discover how to build AI agents with Google's Agent Developer Kit (ADK), exploring workflows, code examples, multi-agent configurations, dynamic memory, and Agent2Agent protocols.

• YouTube
• 39 minutes
• Self-Paced
• Free Video

AI Discovers Physics: Lagrangian and Hamiltonian Neural Networks

Explore how emerging deep learning frameworks like Hamiltonian and Lagrangian neural networks are revolutionizing the discovery of physical laws through automatic differentiation and multi-system data integration.

• YouTube
• 27 minutes
• Self-Paced
• Free Video