Form. Deadline: 16.04, 19:00 MSK
(tentative plan)
| Date | Topic |
|---|---|
| 10.02 | Hyperparameter optimization: SMBO |
| 17.02 | Hyperparameter optimization: gradient-based |
| 24.02 | Technincal meeting 1: project discussion |
| 03.03 | Gaussian processes, SSM |
| 10.03 | Structure selection |
| 17.03 | Technincal meeting 2: proof of concept discussion |
| 24.03 | - |
| 31.03 | Meta-optimization, Genetics, symbolic regression |
| 07.04 | Multi-task learning + Checkpoint |
| 14.04 | Inductive bias |
| 21.04 | Model ensembles |
| 28.04 | Technical meeting 3: overview |
| 05.05 | Latent space projection, kernel emebdding |
| 12.05 | Technical meeting 4: final meeting, final scores |
| Date | Topic |
|---|---|
| 9.09 | Intro |
| 16.09 | Distributions, expectation, likelihood |
| 23.09 | Bayesian inference |
| 30.09 | Technincal meeting 1: project discussion |
| 7.10 | Complexity |
| 14.10 | Var. inference |
| 21.10 | Technincal meeting 2: proof of concept discussion |
| 28.10 | Var. inference 2 |
| 11.11 | - |
| 18.11 | Generative and discriminative models + Checkpoint |
| 25.11 | Graphical models |
| 2.12 | Technical meeting 3: overview |
| 9.12 | Variational inference and optimization |
| 16.12 | Technical meeting 4: final meeting, final scores |
- "Beyond ReLU: how the snake activation fixes neural networks periodic learning problem" by Vladislav Minashkin
- "Gibbs-Based Information Criteria and the Over-Parameterized Regime" by Ivan Papay
- "An Exploration of Softmax Alternatives Belonging to the Spherical Loss Family" by Fedor Sobolevsky
- "Understanding Invariance via Feedforward Inversion of Discriminatively Trained Classifiers" by Dmitrii Vasilenko
- "Bridging Kolmogorov Complexity and Deep Learning" by Sergey Firsov
- "Outlier-Robust Variational Inference: Making Deep Learning More Resilient" by Vladislav Meshkov
- "Accelerated Linearized Laplace Approximation for Bayesian Deep Learning" by Ilya Stepanov
- "Bayesian Neural Networks Under Covariate Shift: When Theory Fails Practice" by Vadim Kasiuk
- "Hybrid Inference: Combining Graphical Models with Graph Neural Networks" by Altay Eynullayev
- "Stochastic Weight Averaging: Finding Wider Optima for Better Generalization" by Gleb Karpeev
- "Hamiltonian Monte Carlo method for sampling" by Denis Rubtsov
- "Distribution Shift in Practice: Strong Baselines vs. Clever Tricks" by Muhammadsharif Nabiev