Lots of Math
20 videos • 104 views • by Tunadorable
Papers that are very math-y
1
Local minima aren't a problem in ML training (Paper Breakdown)
Tunadorable
Download
2
Neural Networks Meet Holographic Brain Theory - FUTURE of AI??? (Paper Breakdown)
Tunadorable
Download
3
Explaining Emergent Abilities in AI (Paper Breakdown)
Tunadorable
Download
4
AI Activation Functions Magnify Infinite-Dimensional Space (Paper Breakdown)
Tunadorable
Download
5
Do Transformers Travel Through an Embedding Vector Landscape? (Paper Breakdown)
Tunadorable
Download
6
Loss Landscapes DO Exhibit Fractal Dynamics (paper breakdown)
Tunadorable
Download
7
Traveling Brainwaves Improve Neural Networks (Paper Breakdown)
Tunadorable
Download
8
Transformer Next-Concept Prediction: Better Than Next-Token Prediction???
Tunadorable
Download
9
Next-Concept Prediction In-depth Math Walkthrough
Tunadorable
Download
10
Meta-Evolution of Intelligent Processes Over Time
Tunadorable
Download
11
Hallucination in Language Models is NEVER Going Away
Tunadorable
Download
12
Let's Code Elon's Grok Model in Pytorch Step-by-Step, From Scratch, Spelled Out
Tunadorable
Download
13
Hierarchical Concept Decoders - A Failed Attempt At Improving GPTs
Tunadorable
Download
14
Geometry and Dynamics of LayerNorm
Tunadorable
Download
15
Underlying Mechanisms Behind Learning Rate Warmup's Success
Tunadorable
Download
16
Accelerated Training by Amplifying Slow Gradients
Tunadorable
Download
17
What happens when you take MoE scaling laws seriously?
Tunadorable
Download
18
What's the difference between 'Inside' OOD and 'Outside' OOD?
Tunadorable
Download
19
[Private video]
Download
20
A new way to compare high dimensional vectors
Tunadorable
Download