Yannic Kilcher Courses

Galactica - A Large Language Model for Science - Drama & Paper Review

Explore Galactica, a language model for science. Learn about its capabilities, controversy, dataset construction, reasoning methods, and performance in various scientific tasks.

• YouTube
• 52 minutes
• On-Demand
• Free Video

Did Google's LaMDA Chatbot Just Become Sentient?

Explore the controversy surrounding Google's LaMDA chatbot and its alleged sentience. Analyze claims, examine language models, and delve into the philosophical questions of AI consciousness and recognition of machine sentience.

• YouTube
• 22 minutes
• On-Demand
• Free Video

Sparse Expert Models - Switch Transformers, GLAM, and More With the Authors

Explore sparse expert models in AI, discussing their architecture, routing, history, applications, and future potential. Learn about Switch Transformers, GLAM, and advancements in efficient large-scale language models.

• YouTube
• 58 minutes
• On-Demand
• Free Video

Typical Decoding for Natural Language Generation - Get More Human-Like Outputs From Language Models

Explore typical sampling, a new decoding method for language models that balances high-probability and high-information outputs, resulting in more diverse and interesting generated text.

• YouTube
• 49 minutes
• On-Demand
• Free Video

One Model for All the Tasks - BLIP - Author Interview

Interview exploring BLIP, a versatile vision-language model that excels in both understanding and generation tasks. Discusses data bootstrapping, architecture evolution, and future directions in multi-modal AI research.

• YouTube
• 49 minutes
• On-Demand
• Free Video

OpenAI Tackles Math - Formal Mathematics Statement Curriculum Learning

Explores OpenAI's innovative approach to formal mathematics using expert iteration, demonstrating improved performance in automated proof search and solving challenging olympiad-level problems.

• YouTube
• 51 minutes
• On-Demand
• Free Video

Competition-Level Code Generation with AlphaCode - Paper Review

Explore DeepMind's AlphaCode system for solving competitive programming problems, its architecture, and performance in real competitions. Learn about key components and innovative techniques.

• YouTube
• 45 minutes
• On-Demand
• Free Video

GPT-NeoX-20B - Open-Source Huge Language Model by EleutherAI - Interview With Co-Founder Connor Leahy

Interview exploring EleutherAI's GPT-NeoX-20B, a 20B parameter open-source language model. Discusses training process, hardware acquisition, model capabilities, and how to try it out.

• YouTube
• 20 minutes
• On-Demand
• Free Video

Player of Games - All the Games, One Algorithm!

Explore a groundbreaking AI algorithm that excels in both perfect and imperfect information games, unifying approaches for chess, Go, poker, and more through guided search, self-play learning, and game theory.

• YouTube
• 54 minutes
• On-Demand
• Free Video

GLIDE- Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models

Explore GLIDE, a text-guided diffusion model for photorealistic image generation and editing. Learn about diffusion models, conditional generation, and guided diffusion techniques for high-quality image synthesis.

• YouTube
• 42 minutes
• On-Demand
• Free Video

Implicit MLE- Backpropagating Through Discrete Exponential Family Distributions

Explore backpropagation through discrete exponential family distributions, enabling integration of combinatorial optimizers in neural networks for enhanced learning capabilities.

• YouTube
• 59 minutes
• On-Demand
• Free Video

Grokking - Generalization Beyond Overfitting on Small Algorithmic Datasets

Explore the fascinating "grokking" phenomenon in neural networks, where sudden pattern recognition leads to perfect generalization. Discover its implications for deep learning and algorithmic datasets.

• YouTube
• 30 minutes
• On-Demand
• Free Video

How Far Can We Scale Up? Deep Learning's Diminishing Returns

Explores the limits of scaling up deep learning, discussing computational costs, CO2 emissions, and potential solutions to overcome diminishing returns in AI model performance.

• YouTube
• 20 minutes
• On-Demand
• Free Video

Inconsistency in Conference Peer Review: Revisiting the 2014 NeurIPS Experiment

Análisis detallado del experimento de revisión por pares de NeurIPS 2014, examinando la subjetividad en las evaluaciones y la capacidad de los revisores para predecir el impacto futuro de los artículos.

• YouTube
• 26 minutes
• On-Demand
• Free Video

Self-Driving from Vision Only - Tesla's Self-Driving Progress by Andrej Karpathy - Talk Analysis

Analysis of Tesla's vision-only self-driving progress, discussing data labeling, edge-case sampling, neural network training, and the advantages of camera-only systems over multi-sensor approaches.

• YouTube
• 24 minutes
• On-Demand
• Free Video