Convolutional Differentiable Logic Gate Networks - NeurIPS Oral - difflogic
14:31
La solidez y completitud de la lógica
13:20
What Are Neural Networks Even Doing? (Manifold Hypothesis)
20:18
Why Does Diffusion Work Better than Auto-Regression?
12:16
How to combine classic algorithms and AI - Learning with Algorithmic Supervision via Relaxations
5:55
Kernel Size and Why Everyone Loves 3x3 - Neural Network Convolution
24:09
The Dark Matter of AI [Mechanistic Interpretability]
26:45
Elegant Geometry of Neural Computations
24:37