2,094,443 Hits in 4.4 sec

Value Prediction Network [article]

Junhyuk Oh, Satinder Singh, Honglak Lee
2017 arXiv   pre-print
This paper proposes a novel deep reinforcement learning (RL) architecture, called Value Prediction Network (VPN), which integrates model-free and model-based RL methods into a single neural network.  ...  In contrast to typical model-based RL methods, VPN learns a dynamics model whose abstract states are trained to make option-conditional predictions of future values (discounted sum of rewards) rather than  ...  Value Prediction Network The value prediction network is developed for semi-Markov decision processes (SMDPs).  ... 
arXiv:1707.03497v2 fatcat:fv2gkdvgrbccrop7qaiwewifuu

PIVEN: A Deep Neural Network for Prediction Intervals with Specific Value Prediction [article]

Eli Simhayev, Gilad Katz, Lior Rokach
2021 arXiv   pre-print
We present PIVEN, a deep neural network for producing both a PI and a value prediction.  ...  Deep learning-based approaches aim to achieve this goal either by improving their prediction of specific values (i.e., point prediction), or by producing prediction intervals (PIs) that quantify uncertainty  ...  We propose PIVEN (prediction intervals with specific value prediction), a novel approach for simultaneous PI generation and value prediction using DNNs.  ... 
arXiv:2006.05139v3 fatcat:bta3zif2vbbjtaq3zro2lqbuq4

Stock market value prediction using neural networks

Mahdi Pakdaman Naeini, Hamidreza Taremian, Homa Baradaran Hashemi
2010 2010 International Conference on Computer Information Systems and Industrial Management Applications (CISIM)  
In this paper, two kinds of neural networks, a feed forward multi layer Perceptron (MLP) and an Elman recurrent network, are used to predict a company's stock value based on its stock share value history  ...  The experimental results show that the application of MLP neural network is more promising in predicting stock value changes rather than Elman recurrent network and linear regression method.  ...  CONCLUSIONS In this paper we used neural networks model to predict the value of stock share in the next day using the previous data about stock market value.  ... 
doi:10.1109/cisim.2010.5643675 dblp:conf/cisim/NaeiniTH10 fatcat:kzpuugjbxbd7lc44cyomdaqp3a

Spatial Correlation and Value Prediction in Convolutional Neural Networks

Gil Shomron, Uri Weiser
2019 IEEE computer architecture letters  
To reduce the number of MACs in CNNs, we propose a value prediction method that exploits the spatial correlation of zero-valued activations within the CNN output feature maps, thereby saving convolution  ...  Convolutional neural networks (CNNs) are a widely used form of deep neural networks, introducing state-of-the-art results for different problems such as image classification, computer vision tasks, and  ...  INTRODUCTION Convolutional neural networks (CNNs) are a widely used form of deep neural networks (DNNs).  ... 
doi:10.1109/lca.2018.2890236 fatcat:mgdkgl56urcfvoch4ncmn4mdfu

Thanks for Nothing: Predicting Zero-Valued Activations with Lightweight Convolutional Neural Networks [article]

Gil Shomron, Ron Banner, Moran Shkolnik, Uri Weiser
2020 arXiv   pre-print
Inspired by the observation that spatial correlation exists in CNN output feature maps (ofms), we propose a method to dynamically predict whether ofm activations are zero-valued or not according to their  ...  Convolutional neural networks (CNNs) introduce state-of-the-art results for various tasks with the price of high computational demands.  ...  convolutional neural network; and finally, the predicted non-zero-valued activations are computed while the zero-valued activations are skipped, thereby saving entire convolution operations ( Figure  ... 
arXiv:1909.07636v3 fatcat:4f4vh6y6o5at3a7c7zwbxxwsr4

Predictable Risks and Predictive Regression in Present-Value Models

Ilaria Piatti, Fabio Trojani
2012 Social Science Research Network  
present-value constraints.  ...  The model implies (i) predictive regressions consistent with a weak return predictability and a missing dividend predictability by aggregate price-dividend ratios, (ii) predictable market volatilities,  ...  value, resulting in less evidence for return predictability.  ... 
doi:10.2139/ssrn.1786897 fatcat:mlobdnuz75burlfwqedofqrohm

An Evaluation of the Predictive Value of Bank Fair Values

Oludimu Oluseun Ehalaiye
2010 Social Science Research Network  
do have predictive value.  ...  The Value-relevance of Fair values -The Predictive ability approach Predictive value is a desirable attribute of an asset (FASB 2010:17) .  ... 
doi:10.2139/ssrn.1716471 fatcat:2xqjx5aywrf2ncg5vfp53rkjw4

The Prediction Value

Maurice Koster, Sascha Kurz, Ines Lindner, Stefan Napel
2013 Social Science Research Network  
The Prediction Value Koster, M.A.L.; Kurz, S.; Lindner, I.; Napel, S.  ...  Abstract We introduce the prediction value (PV) of player i as the difference between the conditional expectations of v(S) when i cooperates or not in a probabilistic TU game.  ...  value We will now provide an axiomatic characterization of the prediction value.  ... 
doi:10.2139/ssrn.2361444 fatcat:v3td43oofzeh7bgcjamnrnztky

Robust Value at Risk Prediction

Loriano Mancini, Fabio Trojani
2010 Social Science Research Network  
This paper proposes a robust semiparametric bootstrap method to estimate predictive distributions of GARCH-type models.  ...  A Monte Carlo simulation shows that our robust method provides more accurate Value at Risk (VaR) forecasts than classical methods, often by a large extent, especially for several days ahead horizons and  ...  Generally, HS and RM methods do not work well, especially for VaR predictions at 1% level with several p-values below 0.05.  ... 
doi:10.2139/ssrn.776124 fatcat:tr7nblreyzef3fkmbpayw5zek4

Measuring the value of accurate link prediction for network seeding

Yijin Wei, Gwen Spencer
2017 Computational Social Networks  
Our contribution: We introduce optimized-against-a-sample (OAS) performance to measure the value of optimizing seeding based on a noisy observation of a network.  ...  Our computational study investigates OAS under several threshold-spread models in synthetic and real-world networks. Our focus is on measuring the value of imprecise link information.  ...  Conclusion Intuitively, as link-prediction error rises, the value of a noisy network observation should decline.  ... 
doi:10.1186/s40649-017-0037-3 pmid:29266137 pmcid:PMC5732613 fatcat:qfkg765rnbdk5d2t72vfrghk4e

Interval-valued Data Prediction via Regularized Artificial Neural Network [article]

Zebin Yang, Dennis K.J. Lin, Aijun Zhang
2018 arXiv   pre-print
A regularized artificial neural network (RANN) is proposed for interval-valued data prediction. The ANN model is selected due to its powerful capability in fitting linear and nonlinear functions.  ...  Experimental results show that the proposed RANN model is an effective tool for interval-valued prediction tasks with high prediction accuracy.  ...  Conclusions This paper proposed a regularized artificial neural network (RANN) model for interval-valued data prediction.  ... 
arXiv:1808.06943v1 fatcat:evu5crimevcablddy6j3kpevje

Link Prediction Based on Graph Topology: The Predictive Value of Generalized Clustering Coefficient

Zan Huang
2010 Social Science Research Network  
This paper represents initial efforts to explore the connection between link prediction and graph topology. The focus is exclusively on the predictive value of the clustering coefficient measure.  ...  These well-studied topological measures and graph generation models have direct implications on link prediction.  ...  Under this stationary assumption, the exact value of observed clustering coefficient should be incorporated to design the link prediction algorithms, such that the predicted links would grow the network  ... 
doi:10.2139/ssrn.1634014 fatcat:v2jery7djrd2leyk7hqjkedjyu

Neural networks for predicting breeding values and genetic gains

Gabi Nunes Silva, Rafael Simões Tomaz, Isabela de Castro Sant'Anna, Moysés Nascimento, Leonardo Lopes Bhering, Cosme Damião Cruz
2014 Scientia Agricola  
After evaluating artificial neural network configurations, our results showed its superiority to estimates based on linear models, as well as its applicability in the genetic value prediction process.  ...  network.  ...  Thus, a direct comparison was made between the measure of network reliability and heritability of the trait, which is the criterion conventionally used for predicting genetic value (maximum likelihood  ... 
doi:10.1590/0103-9016-2014-0057 fatcat:atvykufb3vdizjwkayows2hz4e

Predicting genotypic values associated with gene interactions using neural networks: A simulation study for investigating factors affecting prediction accuracy [article]

Akio Onogi
2019 bioRxiv   pre-print
Neural networks including deep neural networks are attractive candidates to predict phenotypic values.  ...  number of interactions is involved, and (4) neural networks have greater capability to predict epistatic genetic values than random forests, although neural networks are more sensitive to training population  ...  to predict epistatic genetic values.  ... 
doi:10.1101/2019.12.18.881912 fatcat:ctnxnjzysjgspcxny5jl7znrgu

Predicting Target Values of Hydrogen Networks with Purification Unit

A.H. Li, X.F. Wang, Y. Yang, Z.Y. Liu
2017 Chemical Engineering Transactions  
In this paper, when the purification pinch of a hydrogen network involving purification does not change, the target values at a certain purified concentration can be predicted easily and accurately based  ...  on the known target values at original purified concentration.  ...  The main purpose of this paper is to predict the target values of hydrogen networks with fixed purified concentration model.  ... 
doi:10.3303/cet1756081 doaj:87cf2e5dfce84fe4a2bdb08ae353bff9 fatcat:kqao2dfwujhploqvbrtiq4j7qa
« Previous Showing results 1 — 15 out of 2,094,443 results