# Új témák MLJC-re

### Contrastive Learning
* Contrastive Learning Inverts the Data Generating Process - Roland/Yash/Wieland
* CPC
* Contrastive Representation Learning [blogpost](https://lilianweng.github.io/2021/05/31/contrastive-representation-learning.html)

### Meta-learning
(Feri)
* MAML and Reptile \[[1](https://arxiv.org/abs/1703.03400v3), [2](https://arxiv.org/abs/1803.02999v3)\]
* implicit function theorem and iMAML \[[1](https://arxiv.org/abs/1909.04630), [2](https://arxiv.org/abs/1911.02590v1)\]
* Natural Neural Networks and WarpGrad \[[1](https://arxiv.org/abs/1507.00210), [2](https://arxiv.org/abs/1909.00025)\]
* invite: Chelsea Finn

### Natural Gradient Descent
(Feri)
* NGD
* K-FAC and Exact NGD in Linear Networks
* Wasserstein natural gradients
* invite: Michael Arbel

### Differential Privacy
* Algorithmic foundations of Differential Privacy \[[1](https://www.tau.ac.il/~saharon/BigData2018/privacybook.pdf): approx. Chapters 1-3]
* DP-SGD \[[1](https://dl.acm.org/doi/abs/10.1145/2976749.2978318)]
* ?
* invite: ??/Andrew Trask

### Geometric Deep Learning
* Group Equivariant CNNs
* Spherical CNNs
* Gauge Invariant CNNs
* invited speaker: Kondor Risi, Taco Cohen

### Inference in Graphical Models
* Belief Propagation
* Viterbi, forward-backward
* ADF and EM
* Variational Message Passing
* invite: Tom Minka

### Continual Learning
* EWC
* Learning without Forgetting
* Variational Continual Learning
* invite: Raia Hadsell

### PCA Extended
(Patrik)
* Contrastive PCA [paper](https://arxiv.org/abs/1709.06716)
* Probabilistic Contrastive PCA [paper](https://arxiv.org/abs/2012.07977)
* Variational Autoencoders Pursue PCA Directions (by Accident) [paper](https://arxiv.org/abs/1812.06775)

### Reinforcement Learning
* Neu Gergely

### Causal Inference
* [Önreklám - Patrik blogja](https://rpatrik96.github.io/)

### Few-shot/small data

### Domain adaptation

### VAEs & Co
* VAEBM
* PCA-VAE
* iVAE

### EBMs
* MCMC
* Score Matching
* VAEBM
* How to train your EBM

### Network & Data visualization
* Saliency map
* Feature visualization
* Interpretability
* UMAP : [On UMAP's true loss function](http://arxiv.org/abs/2103.14608)
* t-SNE [post](https://distill.pub/2016/misread-tsne/)
* Laplacian eigenmaps
* Locally linear embeddig
* Isomaps