Papers
Papers with Notes:
- Intention Conditioned Value Functions (ICVF)
- Neural Acceleration of Graph Based Utility Functions for Sparse Matrices
- Neural Acceleration for Incomplete Factorization Preconditioning
- Grokking - Generalization Beyond Overfitting on Small Algorithmic Datasets
- Grokfast - Accelerated Grokking by Ampilfying Slow Gradients
- Evaluation of Pre-training LLMs on Leadership Class Supercomputers
- Every Model Learned by Gradient Descent Is Approximately a Kernel Machine
- Converting Neural Networks to Symbolic Models
- Iterative alpha-(de)blending
- Learning to Reach Goals via Iterated Supervised Learning
- Goal-Conditioned Supervised Learning
- Reward Conditioned Policies
- The Surprising Effectiveness of PPO in Cooperative Multi-Agent Games
Minimal notes:
Other articles:
1. Interesting Stuff
1.0.1. Word-As-Image for Semantic Typography
https://wordasimage.github.io/Word-As-Image-Page/
Figure 1: Word As Image
1.0.2. CLIPasso: Semantically-Aware Object Sketching
https://clipasso.github.io/clipasso/
Figure 2: CLIPass