Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

New posts #52

Open
2 of 16 tasks
OleguerCanal opened this issue Nov 1, 2020 · 0 comments
Open
2 of 16 tasks

New posts #52

OleguerCanal opened this issue Nov 1, 2020 · 0 comments
Assignees
Labels

Comments

@OleguerCanal
Copy link
Member

OleguerCanal commented Nov 1, 2020

This is just for me to remember the new posts I want to write about / refactor:

  • Attention-based Models & self-attention
  • Graph NN
  • Dimensionality reduction approaches
  • Autoencoders history
  • GANs history
  • Paper: Analytic Manifold Learning: Unifying & Evaluating Representations for Continuous Control
  • Paper: Phasic Policy Gradient https://arxiv.org/abs/2009.04416
  • What is a matrix? Matrices as data vs Matrices as transformations. + interpretations. Inspiration Inspiration
  • Particle filters vs HMM vs RL vs (Contextual) Bayesian Optimization vs Multi-Bandit Arm
  • How to evaluate models: Common metrics depending on model's nature Precision/Recall Single&Multiple Variables, ROC AUC...
  • Continual learning: Summarize references from Raia Hadsell M2L talk: https://www.sciencedirect.com/science/article/pii/S1364661320302199

Refactor/Review:

  • First three lectures of RL: Change order and unify style
  • Distributions post: Add distributions and meaning, Add KL generalizations, Jensen–Shannon divergence, Total variation distance, Wasserstein metric...
  • Model-free RL
  • Model-based RL: Watch new lectures and try to re-organize writing
  • Compress oleguer profile picture
@OleguerCanal OleguerCanal self-assigned this Nov 1, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

1 participant