What Is ChatGPT Doing? Training Neural Networks - Episode 4
Explore the inner workings of ChatGPT and large language models, covering neural network training, fine-tuning, reinforcement learning, and parameter optimization in this insightful discussion.
GFlowNet Foundations and Applications in Biological Sequence Design
Explore GFlowNet foundations and applications in biological sequence design. Learn about flow conservation, transition probabilities, and practical implementations for molecule generation and peptide design.
Local Search GFlowNets - Enhancing Sampling Quality for Drug Discovery
Explore Local Search GFlowNets, a novel algorithm enhancing sampling quality for structured sequence generation in drug discovery and other fields through improved targeting of high-reward regions.
Applying Active Learning in Drug Discovery - There's No Free Lunch, but You Can Get a Discount
Explore active learning applications in drug discovery, from prioritizing molecules for synthesis to virtual screening of large libraries, with insights on machine learning's impact across the process.
Generation of Verified Assembly Code Using Dafny and Reinforcement Learning
Explore verified assembly code generation using Dafny and reinforcement learning, combining efficiency with automatic correctness verification for complex programming tasks.
Programmatic Strategy Synthesis: Resolving Nondeterminism in Probabilistic Programs
Explore programmatic strategy synthesis for resolving nondeterminism in probabilistic programs using deductive verification techniques and weakest precondition-style frameworks.
Deep Active Learning: Enhancing Data Efficiency in Machine Learning - Lecture
Explore deep learning's data efficiency, active learning in NLP, and innovative approaches to enhance learning from human interactions in AI systems.
Machines that Learn Through Action - The Future of AI
Explore the future of AI with Julie Pitt as she discusses machines that learn through action, focusing on key advancements and potential impacts on various industries.
Learning Agents for Games using Databricks
Accelerate reinforcement learning for games with Databricks' advanced analytics, collaborative tools, and rapid deployment capabilities.
Retrospective on Multi-Agent Learning as a Path to Intelligence
Explore the potential of multi-agent learning in advancing artificial intelligence, drawing insights from psychology and neuroscience perspectives.
How OpenAI Solved the Data Wall Problem - Synthesizing Infinite Provable Data for Reasoning
Explore OpenAI's innovative solution to the data wall problem through synthesizing infinite provable data for enhanced reasoning capabilities.
Generative AI and Large Language Models: Fine-tuning with SageMaker, PEFT, RLHF and PPO
Master advanced LLM fine-tuning techniques using Amazon SageMaker, exploring PEFT, RLHF, and PPO methodologies for optimized AI model performance and customization.
Runaway Feedback Loops in Predictive Policing
Explore runaway feedback loops in predictive policing, examining assumptions, theory, and potential solutions to address reinforcement issues in law enforcement algorithms.
Training an Autonomous Pentester with Deep RL
Explore deep reinforcement learning for autonomous pentesting, using Metasploit and simulated networks to train an AI agent capable of compromising real-world vulnerable systems.
System for Deepfake Detection - DeepAudioDetector - Ekoparty 2021
Explora técnicas avanzadas de detección de deepfakes de audio utilizando inteligencia artificial, con énfasis en metodologías, experimentos y herramientas de libre uso para enfrentar amenazas cibernéticas emergentes.
Adversarial Bandits with Knapsacks
Explore dynamic pricing and bandits with knapsacks in adversarial settings, covering algorithms, challenges, and future research directions in this field.
Minecraft as AI Playground and Laboratory
Explore Minecraft as an AI research platform, focusing on Project Malmo and the MineRL competition to advance sample-efficient reinforcement learning and tackle complex decision-making challenges.
Reinforcement Learning in Action - Creating Arena Battle AI for Blade & Soul
Discover how NCSOFT created pro-level AI agents for arena battles in Blade & Soul using reinforcement learning, addressing complexity and real-time response challenges.
Successfully Use Deep Reinforcement Learning in Testing and NPC Development
Explore deep reinforcement learning for game testing and NPC development, focusing on effective approaches, cost mitigation, and real-world applications in Unity projects.
Ruling StarCraft Game Spitefully - Exploiting the Blind Spot of AI-Powered Game Bots
Explore techniques to exploit vulnerabilities in AI-powered game bots, focusing on StarCraft. Learn about deep reinforcement learning, adversarial attacks, and potential defenses in gaming AI.
Democratize Machine Learning - ML-Agents Explained
Explore Unity ML-Agents: an open-source toolkit bridging Unity and machine learning. Learn about reinforcement and imitation learning techniques for innovative game development and AI applications.