Zobrazeno 1 - 10
of 94 569
pro vyhledávání: '"partial information"'
Self Supervised learning (SSL) has demonstrated its effectiveness in feature learning from unlabeled data. Regarding this success, there have been some arguments on the role that mutual information plays within the SSL framework. Some works argued fo
Externí odkaz:
http://arxiv.org/abs/2412.02121
Autor:
Rao, P Raghavendra, Vyavahare, Pooja
This work studies the distributed learning process on a network of agents. Agents make partial observation about an unknown hypothesis and iteratively share their beliefs over a set of possible hypotheses with their neighbors to learn the true hypoth
Externí odkaz:
http://arxiv.org/abs/2411.11411
Autor:
Dissanayake, Pasan, Hamman, Faisal, Halder, Barproda, Sucholutsky, Ilia, Zhang, Qiuyi, Dutta, Sanghamitra
Knowledge distillation provides an effective method for deploying complex machine learning models in resource-constrained environments. It typically involves training a smaller student model to emulate either the probabilistic outputs or the internal
Externí odkaz:
http://arxiv.org/abs/2411.07483
This paper investigates a linear quadratic stochastic optimal control (LQSOC) problem with partial information. Firstly, by introducing two Riccati equations and a backward stochastic differential equation (BSDE), we solve this LQSOC problem under st
Externí odkaz:
http://arxiv.org/abs/2409.16924
Exploring the data sources used to train Large Language Models (LLMs) is a crucial direction in investigating potential copyright infringement by these models. While this approach can identify the possible use of copyrighted materials in training dat
Externí odkaz:
http://arxiv.org/abs/2409.13831
The framework of Partial Information Decomposition (PID) unveils complex nonlinear interactions in network systems by dissecting the mutual information (MI) between a target variable and several source variables. While PID measures have been formulat
Externí odkaz:
http://arxiv.org/abs/2409.13506
We consider a central trading desk which aggregates the inflow of clients' orders with unobserved toxicity, i.e. persistent adverse directionality. The desk chooses either to internalise the inflow or externalise it to the market in a cost effective
Externí odkaz:
http://arxiv.org/abs/2407.04510