Yannic Kilcher Courses

Active Dendrites Avoid Catastrophic Forgetting - Interview With the Authors

Interview exploring how active dendrites in neural networks can mitigate catastrophic forgetting, enabling better multi-task and continual learning for AI systems in dynamic environments.

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

Avoiding Catastrophe - Active Dendrites Enable Multi-Task Learning in Dynamic Environments

Explore how active dendrites in neural networks mitigate catastrophic forgetting, enabling multi-task learning and adaptation in dynamic environments through context-sensitive gating and sparse representations.

• YouTube
• 1 hour 5 minutes
• Self-Paced
• Free Video

Author Interview - VOS- Learning What You Don't Know by Virtual Outlier Synthesis

Interview exploring Virtual Outlier Synthesis for out-of-distribution detection in neural networks. Discusses motivations, methodology, and key findings of novel approach to improve model robustness using synthesized outlier data.

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

Learning What You Don't Know by Virtual Outlier Synthesis - Paper Explained

Explore Virtual Outlier Synthesis, a novel method for improving out-of-distribution detection in deep learning models by generating synthetic outliers during training to enhance decision boundary regularization.

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

Spurious Normativity Enhances Learning of Compliance and Enforcement Behavior in Artificial Agents

Explore how arbitrary social norms enhance learning of rule enforcement in AI agents, with insights on societal implications and potential applications in artificial general intelligence.

• YouTube
• 1 hour 37 minutes
• Self-Paced
• Free Video

AI and Formal Math - Formal Mathematics Statement Curriculum Learning

Explore AI's potential in formal mathematics through expert iteration, curriculum learning, and language modeling. Discover how machines and humans can collaborate to solve complex mathematical problems.

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

AlphaCode - With the Authors

Explore AlphaCode's development, capabilities, and future potential in competitive programming through an in-depth interview with its creators, covering technical aspects and real-world applications.

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

Can Wikipedia Help Offline Reinforcement Learning - Author Interview

Exploring the potential of pre-trained language models in offline reinforcement learning, discussing experimental results, model performance, and future directions for leveraging sequence modeling techniques in RL tasks.

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

Can Wikipedia Help Offline Reinforcement Learning? - Paper Explained

Exploring how pre-training on Wikipedia improves offline reinforcement learning models, enhancing performance, reducing parameters, and revealing connections between language and RL domains.

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

All About AI Accelerators - GPU, TPU, Dataflow, Near-Memory, Optical, Neuromorphic & More

Comprehensive exploration of AI accelerator technologies, from GPUs to emerging innovations like neuromorphic computing, discussing their principles, advantages, and future potential in advancing artificial intelligence.

• YouTube
• 1 hour 3 minutes
• Self-Paced
• Free Video

CM3- A Causal Masked Multimodal Model of the Internet

Explore CM3, a groundbreaking multimodal model that processes HTML, text, and images. Learn about its innovative training strategy and potential applications in various AI tasks.

• YouTube
• 1 hour 24 minutes
• Self-Paced
• Free Video

HyperTransformer - Model Generation for Supervised and Semi-Supervised Few-Shot Learning

Explore HyperTransformer, a novel approach to few-shot learning using transformers to generate CNN weights. Learn about its architecture, advantages, and potential applications in machine learning.

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

Language Models as Zero-Shot Planners - Extracting Actionable Knowledge for Embodied Agents

Exploring how large language models can translate high-level tasks into actionable steps for virtual environments, with techniques to improve executability and potential real-world applications.

• YouTube
• 1 hour 17 minutes
• Self-Paced
• Free Video

Unsupervised Brain Models - How Does Deep Learning Inform Neuroscience?

Explore connections between deep learning and neuroscience, focusing on unsupervised brain models and their potential to explain visual processing streams and representation learning in the brain.

• YouTube
• 1 hour 21 minutes
• Self-Paced
• Free Video

Predicting the Rules Behind - Deep Symbolic Regression for Recurrent Sequences

Explore deep symbolic regression for predicting rules behind number sequences using transformers. Learn about data encoding, training processes, and applications beyond mathematics in this in-depth interview.

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