Provably Convergent Policy Optimization via Metric-aware Trust Region Methods

Jun Song, Niao He, Lijun Ding, Chaoyue Zhao

Action editor: Amir-massoud Farahmand.

https://openreview.net/forum?id=jkTqJJOGMS

#optimality #reinforcement #lagrangian

Expectation of the maximum of Normal random variables with applications to reinforcement learning

https://openreview.net/forum?id=ZvM6TVPBBM

#reinforcement #expectation #optimality

'Preconditioned Gradient Descent for Overparameterized Nonconvex Burer--Monteiro Factorization with Global Optimality Certification', by Gavin Zhang, Salar Fattahi, Richard Y. Zhang.

http://jmlr.org/papers/v24/22-0882.html

#optimality #minimizer #overparameterization

Provably Convergent Policy Optimization via Metric-aware Trust Region Methods

https://openreview.net/forum?id=jkTqJJOGMS

#optimality #reinforcement #lagrangian

'Reinforcement Learning for Joint Optimization of Multiple Rewards', by Mridul Agarwal, Vaneet Aggarwal.

http://jmlr.org/papers/v24/19-980.html

#rewards #reinforcement #optimality

Defense Against Reward Poisoning Attacks in Reinforcement Learning

Kiarash Banihashem, Adish Singla, Goran Radanovic

https://openreview.net/forum?id=goPsLn3RVo

#reinforcement #reward #optimality

I dislike modulo 2 addition in linear error correction codes. I prefer the monotone Boolean circuit instead. NOT operations are trouble makers. Without them life is much better. Therefore I used a monotone circuit in my deep learning model. https://vixra.org/abs/2212.0193
#deeplearning #monotonecircuit #ErrorCorrectionCode #optimality

New #openaccess #paper: "Vegetation optimality explains the convergence of catchments on the Budyko curve", published today in HESS by Remko Nijzink and myself: https://hess.copernicus.org/articles/26/6289/2022/
The paper also follows a strict #openscience approach, check it out!
#research #science #hydrology #vegetation #optimality @list_hydrocat

#introduction Hello everybody! I am trying to do research in computational biology (both in systems and synthetic bio). Big fan of #optimality principles in biology. Part of the #twittermigration. Looking forward to re-building my network, and expanding it, in this new place. Still trying to figure Mastodon-related things out...