Filters








733,313 Hits in 4.3 sec

Learning Populations of Parameters [article]

Kevin Tian, Weihao Kong, Gregory Valiant
2017 arXiv   pre-print
We also extend our results to the multi-dimensional parameter case, capturing settings where each member of the population has multiple associated parameters.  ...  sports analytics, and variation in the gender ratio of offspring.  ...  Acknowledgments We thank Kaja Borge and Ane Nødtvedt for sharing an anonymized dataset on sex composition of dog litters, based on data collected by the Norwegian Kennel Club.  ... 
arXiv:1709.02707v2 fatcat:s6yap3h4ancahbjp5p66ovjbli

Learning the Structure and Parameters of Large-Population Graphical Games from Behavioral Data [article]

Jean Honorio, Luis Ortiz
2015 arXiv   pre-print
We consider learning, from strictly behavioral data, the structure and parameters of linear influence games (LIGs), a class of parametric graphical games introduced by Irfan and Ortiz (2014).  ...  Motivated by the computational work on LIGs, we cast the learning problem as maximum-likelihood estimation (MLE) of a generative model defined by pure-strategy Nash equilibria (PSNE).  ...  We warmly thank Tommi Jaakkola for several informal discussions on the topic of learning games and the ideas behind causal strategic inference, as well as suggestions on how to improve the presentation  ... 
arXiv:1206.3713v4 fatcat:xocfcdz44fec3m5uqj5zhlivze

Maximum Likelihood Estimation for Learning Populations of Parameters [article]

Ramya Korlakai Vinayak, Weihao Kong, Gregory Valiant, Sham M. Kakade
2019 arXiv   pre-print
This problem arises in numerous domains, including the social sciences, psychology, health-care, and biology, where the size of the population under study is usually large while the number of observations  ...  Consider a setting with N independent individuals, each with an unknown parameter, p_i ∈ [0, 1] drawn from some unknown distribution P^.  ...  We use the lens of sparse regime analysis for this problem of learning a population of parameters.  ... 
arXiv:1902.04553v1 fatcat:qfds4fg4xzhffbshrqmyli2maq

Automated Determination of Stellar Population Parameters in Galaxies Using Active Instance-based Learning [article]

Thamar Solorio, Elena Terlevich Osservatorio Astronomico di Padova, Italy)
2003 arXiv   pre-print
Experimental results show that this method can estimate with high speed and accuracy the physical parameters of the stellar populations.  ...  In this work we focus on the determination of the relative distributions of young, intermediate-age and old populations of stars in galaxies.  ...  to estimate the reddening parameters of the three populations.  ... 
arXiv:astro-ph/0312073v1 fatcat:opam7moenfcargnvehj5uy4dyq

Parameter as a Switch Between Dynamical States of a Network in Population Decoding

Jiali Yu, Hua Mao, Zhang Yi
2017 IEEE Transactions on Neural Networks and Learning Systems  
Population coding is a method to represent stimuli using the collective activities of a number of neurons.  ...  Moreover, it is a challenge to identify the right parameter of the decoding model, which plays a key role for convergence.  ...  Therefore, the value of the parameter is crucial for decoding the population activities. VII.  ... 
doi:10.1109/tnnls.2015.2485263 pmid:26513809 fatcat:7esfvhtvmrejncvgxqn3xrypji

Mathematical derivations and supplemental figures from The importance of life history and population regulation for the evolution of social learning

Dominik Deffner, Richard McElreath
2020 Figshare  
Social learning and life history interact in human adaptation, but nearly all models of the evolution of social learning omit age structure and population regulation.  ...  We discuss why life history and age structure are important for social learning and present an exemplary model of the evolution of social learning in which demographic properties of the population arise  ...  Results are averaged over all values of other parameters. Figure S5 . S5 Results for simulations with low cost of individual learning (c = 0.01).  ... 
doi:10.6084/m9.figshare.12279650.v1 fatcat:j2np4e3irbbfteitedy7grrnwm

Evolutionary Architecture Search for Graph Neural Networks [article]

Min Shi, David A.Wilson, Xingquan Zhu, Yu Huang, Yuan Zhuang, Jianxun Liu, Yufei Tang
2020 arXiv   pre-print
In addition, the slight variation of hyper parameters such as learning rate and dropout rate could dramatically hurt the learning capacity of GNN.  ...  Instead of optimizing only the model structures with fixed parameter settings as existing work, an alternating evolution process is performed between GNN structures and learning parameters to dynamically  ...  Then, the population (P 0 ) of GNN parameters w.r.t S 0 is initialized and evolved to identify the optimal parameter setting (e.g., learning rate and dropout rate).  ... 
arXiv:2009.10199v1 fatcat:k2h23byz2jfebbiw3mzckveblm

Population rule learning in symmetric normal-form games: theory and evidence

Dale O. Stahl
2001 Journal of Economic Behavior and Organization  
When predicting the population distribution of choices and accounting for the number of parameters, the population rule learning model is much better than aggregation of individually estimated rule learning  ...  A model of population rule learning is formulated and estimated using experimental data.  ...  Ray Battalio assisted in the design of the computer interface, and Ernan Haruvy provided research assistance. However, all errors and omissions are the sole responsibility of the author.  ... 
doi:10.1016/s0167-2681(00)00169-4 fatcat:nea3q2fknffdpgdzihfwirp4jm

Hybrid Evolutionary Learning Approaches for The Virus Game

M.H. Naveed, P.I. Cowling, M.A. Hossain
2007 2007 IEEE Symposium on Computational Intelligence and Games  
This paper investigates the effectiveness of hybrids of learning and evolutionary approaches to find weights and topologies for an artificial neural network (ANN) which is used to evaluate board positions  ...  The results show that evolutionary RPROP and evolutionary BP have significantly better generalisation performance than their constituent learning and evolutionary methods.  ...  This approach differs from the hybrid methods proposed by [4] and [6] in that two different populations (of learning parameters and network topologies) are maintained and Π = Population of RPROP parameter  ... 
doi:10.1109/cig.2007.368098 dblp:conf/cig/NaveedCH07 fatcat:5bdrhb6jzrde5itrbfwgzdgdnm

Self-adaptive parameters in genetic algorithms

Eric Pellerin, Luc Pigeon, Sylvain Delisle, Belur V. Dasarathy
2004 Data Mining and Knowledge Discovery: Theory, Tools, and Technology VI  
Our preliminary results show that a GA is able to learn and evaluate the quality of self-set parameters according to their degree of contribution to the resolution of the problem.  ...  The problem of interest to us here is the self-adaptive parameters adjustment of a GA.  ...  One of the potential solutions to determine the best set of parameters resides in the use of learning. This is the problem we consider here: the learning-based self-adjustment of a GA's parameters.  ... 
doi:10.1117/12.542156 dblp:conf/dmkdttt/PellerinPD04 fatcat:to5bxvelmvc5lobk4s6phj3edi

Parameter Adaptation within Co-adaptive Learning Classifier Systems [chapter]

Chung-Yuan Huang, Chuen-Tsai Sun
2004 Lecture Notes in Computer Science  
The authors propose a co-adaptive approach to controlling parameters for coevolution-based learning classifier systems.  ...  The system combines the advantages of both adaptive and self-adaptive parameter-control approaches.  ...  As one would expect, the population size performance metric (defined as the number of macro-classifiers in a classifier population) belongs to the category of population state measures.  ... 
doi:10.1007/978-3-540-24855-2_92 fatcat:xxzbwm3hvrcsjcog7ikgu2sheu

Importance of Parameter Settings on the Benefits of Robot-to-Robot Learning in Evolutionary Robotics

Jacqueline Heinerman, Evert Haasdijk, A. E. Eiben
2019 Frontiers in Robotics and AI  
First, robot-to-robot learning can reduce the number of bad performing individuals in the population.  ...  As a result, we show that this type of social learning can reduce the sensitivity of the learning process to the choice of parameters in two ways.  ...  PARAMETER SETTINGS AND MEASUREMENTS Our implementation of the individual learning process has 21 parameters. These parameters are related to the individual learning mechanism of one robot.  ... 
doi:10.3389/frobt.2019.00010 pmid:33501027 pmcid:PMC7806056 fatcat:ydrgm33slnfe7jyi2vd7l5ec7y

The Evolutionary Dynamics of Independent Learning Agents in Population Games [article]

Shuyue Hu, Chin-Wing Leung, Ho-fung Leung, Harold Soh
2020 arXiv   pre-print
the expected behaviour of a population.  ...  In addition, we present extensive experimental results validating that Theorem 1 holds for a variety of learning methods and population games.  ...  Note that the density ppx, tq is intuitively the proportion of agents having the vector x of critical parameters in the population.  ... 
arXiv:2006.16068v1 fatcat:ljcaz7l2mfdnhco4daanc5tkiq

Using evolution to improve neural network learning: pitfalls and solutions

John A. Bullinaria
2007 Neural computing & applications (Print)  
The use of evolutionary techniques to improve the learning abilities of neural network systems is now widespread.  ...  Autonomous neural network systems typically require fast learning and good generalization performance, and there is potentially a trade-off between the two.  ...  ACKNOWLEDGEMENTS This paper brings together into a unified framework, and expands upon, a number of ideas and results previously scattered across several conference papers [12, 15, 17, 22, 33, 34] .  ... 
doi:10.1007/s00521-007-0087-9 fatcat:i7nvztxegveptl5adjm437dj2a

Is social learning more than parameter tuning?

Jacqueline Heinerman, Jörg Stork, Margarita Alejandra Rebolledo Coy, Julien Hubert, A. E. Eiben, Thomas Bartz-Beielstein, Evert Haasdijk
2017 Proceedings of the Genetic and Evolutionary Computation Conference Companion on - GECCO '17  
population size from the 1 robot setup divided by the number of robots (e.g., when the 1 robot setup has a population size of 100, the social learning experiments used a population size of 50 and 25 for  ...  When social learning is applied, these robots have the same parameter se ings as the individual learning mechanisms except for the population size. e population size for the 2 and 4 robot setup is the  ... 
doi:10.1145/3067695.3076059 dblp:conf/gecco/HeinermanSCHEBH17 fatcat:t3ftbxz435bdjpm3g5p4m7va6i
« Previous Showing results 1 — 15 out of 733,313 results