Poster Session
Abstract
Unrolling versus bilevel optimization in the context of learning variational models
Niklas BreustedtTU Braunschweig, Germany
Please see the abstract as PDF file.
Online Adaptive Learning in Energy Trading Stackelberg Games with Time-Coupling Constraints
Styliani KampezidouGeorgia Institute of Technology (Atlanta), USA
An energy trading mechanism is proposed between a selfish energy broker (aggregator) and her selfish energy customers (prosumers). The proposed design is a Stackelberg game where the aggregator trades energy bidirectionally between the prosumers and the wholesale electricity market for profit. For the purpose of satisfying the prosumers' desired energy consumption, time-coupling constraints are introduced. The described game does not admit closed form equilibrium strategies and therefore an online adaptive learning algorithm is proposed to mitigate this challenge. The latter is scalable with the number of prosumers and does not require explicit knowledge of the prosumers' decision-making mechanisms. Experimental results that utilize real-world data from the California market are provided to demonstrate the performance of the proposed approach.
Taming neural networks with TUSLA: Non-convex learning via adaptive stochastic gradient Langevin algorithms
Iosif LytrasUniversity of Edinburgh, United Kingdom
Joint work with Attila Lovas, Miklós Rásonyi and Sotirios Sabanis
Arxiv Url : arxiv.org/pdf/2006.14514.pdf
Please see the abstract as PDF file.
The geometry of discounted stationary distributions of Markov decision processes
Johannes MüllerMax Planck Institute for Mathematics in the Sciences (Leipzig), Germany
Please see the abstract as PDF file.
Investment Vs. reward in competitive games
Oren NeumannGoethe University Frankfurt am Main, Germany
Please see the abstract as PDF file.
PAC-Bayesian Estimation for High-Dimensional Multi-Index Regression with Unknown Active Dimension
Maximilian SteffenUniversität Hamburg, Germany
Please see the abstract as PDF file.
Adversarial Perturbation Stability of the Layered Group Basis Pursuit
David SzeghyEötvös Loránd University (ELTE) (Budapest), Hungary
Please see the abstract as PDF file.
On the Expected Complexity of Maxout Networks
Hanna TseranMPI for Mathematics in the Sciences (Leipzig), Germany
Please see the abstract as PDF file.
Parametrisation Independence of the Natural Gradient in Overparametrised Systems
Jesse van OostrumHamburg University of Technology, Germany
Please see the abstract as PDF file.
Natural Reweighted Wake Sleep for Convolutional Networks
Csongor-Huba VaradyMax Planck Institute for Mathematics in the Sciences (MiS) in Leipzig., Germany
Please see the abstract as PDF file.
Lifted Bregman Networks
Xiaoyu WangUniversity of Cambridge, United Kingdom
Please see the abstract as PDF file.
Applications of associative algebras in machine learning
Chia ZargehUniversity of Sao Paulo, Brazil
Please see the abstract as PDF file.