Zobrazeno 1 - 10
of 163
pro vyhledávání: '"Velarde, Pedro"'
Autor:
Cisneros-Velarde, Pedro
Social balance is a concept in sociology which states that if every three individuals in a population achieve certain structures of positive or negative interactions, then the whole population ends up in one faction of positive interactions or divide
Externí odkaz:
http://arxiv.org/abs/2410.04054
Weight normalization (WeightNorm) is widely used in practice for the training of deep neural networks and modern deep learning libraries have built-in implementations of it. In this paper, we provide the first theoretical characterizations of both op
Externí odkaz:
http://arxiv.org/abs/2409.08935
Autor:
Cisneros-Velarde, Pedro
We study the evolution of opinions inside a population of interacting large language models (LLMs). Every LLM needs to decide how much funding to allocate to an item with three initial possibilities: full, partial, or no funding. We identify biases t
Externí odkaz:
http://arxiv.org/abs/2406.15492
Autor:
Cisneros-Velarde, Pedro, Koyejo, Sanmi
Nash Q-learning may be considered one of the first and most known algorithms in multi-agent reinforcement learning (MARL) for learning policies that constitute a Nash equilibrium of an underlying general-sum Markov game. Its original proof provided a
Externí odkaz:
http://arxiv.org/abs/2303.00177
We consider the problem of optimization of deep learning models with smooth activation functions. While there exist influential results on the problem from the ``near initialization'' perspective, we shed considerable new light on the problem. In par
Externí odkaz:
http://arxiv.org/abs/2209.15106
Autor:
Ren, Shenyuan, Shi, Yuanfeng, Berg, Quincy Y. van den, Firmansyah, Muhammad, Chung, Hyun-Kyung, Fernandez-Tello, Elisa V., Velarde, Pedro, Wark, Justin S., Vinko, Sam M.
The advent of x-ray free-electron lasers (XFELs) has enabled a range of new experimental investigations into the properties of matter driven to extreme conditions via intense x-ray-matter interactions. The femtosecond timescales of these interactions
Externí odkaz:
http://arxiv.org/abs/2208.00573
While the difficulty of reinforcement learning problems is typically related to the complexity of their state spaces, Abstraction proposes that solutions often lie in simpler underlying latent spaces. Prior works have focused on learning either a con
Externí odkaz:
http://arxiv.org/abs/2206.03467
Although parallelism has been extensively used in reinforcement learning (RL), the quantitative effects of parallel exploration are not well understood theoretically. We study the benefits of simple parallel exploration for reward-free RL in linear M
Externí odkaz:
http://arxiv.org/abs/2205.15891
Recently, a special case of precision matrix estimation based on a distributionally robust optimization (DRO) framework has been shown to be equivalent to the graphical lasso. From this formulation, a method for choosing the regularization term, i.e.
Externí odkaz:
http://arxiv.org/abs/2201.12441
The design of fixed point algorithms is at the heart of monotone operator theory, convex analysis, and of many modern optimization problems arising in machine learning and control. This tutorial reviews recent advances in understanding the relationsh
Externí odkaz:
http://arxiv.org/abs/2110.03623