Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

The objective of this master thesis is to find transfer learning
methods which can be applied in reinforcement learning problems.
This method has to improves the learning speed of an agent on a
target task, given a set of already trained policies on similar
simpler tasks. We used Q learning algorithm combined to neural
network to represent Q function that determine how an agent act
on it’s assigned task.
Using source network trained on simpler problems, we manage to
find an architecture that combines networks with an interface to
transfer knowledge in the new network. This technique allowed us
to solve the OpenAI Gym mountain car problem using source
networks trained on a smaller mountain and a faster car.
We tested more complex connection architectures to improve the
representation complexity of the combined networks.
We also implemented a method to learn multiple networks in
parallel to exploit the ensemble properties when learning a new
task. This method improves the convergence and results.
Finally we tested our implementation on the OpenAI Gym
Cartpole environment with succes.

transfer --- learning --- mountain --- car --- cartpole --- neural --- network --- concatenation --- merge --- openai --- gym --- reinforcement --- Ingénierie, informatique & technologie > Sciences informatiques

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

'Practical Microsimulation Modelling' brings together a description and examples of the main methods used in microsimulation modelling used in the field of income distribution analysis. It is structured to develop and use the different types of models used in the field, with a focus on household targeted policy.

Econometric models. --- Statistical matching. --- Appariement (Statistique) --- Income distribution --- Mathematical models. --- Concatenation, File (Statistics) --- Data fusion (Statistics) --- Data matching (Statistics) --- Data merging (Statistics) --- File concatenation (Statistics) --- Fusion, Data (Statistics) --- Imputation, Mass (Statistics) --- Mass imputation (Statistics) --- Matching, Data (Statistics) --- Matching, Statistical --- Merging, Data (Statistics) --- Microsimulation modeling (Statistics) --- Modeling, Microsimulation (Statistics) --- Sampling (Statistics) --- Econometric models --- Statistical matching

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

This handbook brings together the latest research on applied market design. It surveys matching markets: environments where there is a need to match large two-sided populations to one another, such as law clerks and judges or patients and kidney donors.

Game theory. --- Markets --- Statistical matching. --- Supply and demand. --- Mathematical models. --- Demand and supply --- Industrial production --- Law of supply and demand --- Economics --- Competition --- Exchange --- Overproduction --- Prices --- Value --- Concatenation, File (Statistics) --- Data fusion (Statistics) --- Data matching (Statistics) --- Data merging (Statistics) --- File concatenation (Statistics) --- Fusion, Data (Statistics) --- Imputation, Mass (Statistics) --- Mass imputation (Statistics) --- Matching, Data (Statistics) --- Matching, Statistical --- Merging, Data (Statistics) --- Microsimulation modeling (Statistics) --- Modeling, Microsimulation (Statistics) --- Sampling (Statistics) --- Public markets --- Commerce --- Fairs --- Market towns --- Games, Theory of --- Theory of games --- Mathematical models --- Mathematics --- Supply and demand --- Statistical matching --- Game theory --- E-books

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Verena Puchner evaluates and compares statistical matching and selected SAE methods. Due to the fact that poverty estimation at regional level based on EU-SILC samples is not of adequate accuracy, the quality of the estimations should be improved by additionally incorporating micro census data. The aim is to find the best method for the estimation of poverty in terms of small bias and small variance with the aid of a simulated artificial "close-to-reality" population. Variables of interest are imputed into the micro census data sets with the help of the EU-SILC samples through regression models including selected unit-level small area methods and statistical matching methods. Poverty indicators are then estimated. The author evaluates and compares the bias and variance for the direct estimator and the various methods. The variance is desired to be reduced by the larger sample size of the micro census.  Contents Regression Models Including Selected Small Area Methods Statistical Matching Application to Poverty Estimation Using EU-SILC and Micro Census Data Bootstrap Methods Target Groups  Researchers, students, and practitioners in the fields of statistics, official statistics, and survey statistics  The Author Verena Puchner obtained her master’s degree at Technical University of Vienna under the supervision of Priv.-Doz. Dipl.-Ing. Dr. techn. Matthias Templ. At present, she works as a data miner and consultant.

Mathematics. --- Computational Mathematics and Numerical Analysis. --- Probability Theory and Stochastic Processes. --- Applications of Mathematics. --- Computer science --- Distribution (Probability theory). --- Mathématiques --- Informatique --- Distribution (Théorie des probabilités) --- Computer science -- Mathematics. --- Poverty -- Statistical methods. --- Mathematics --- Physical Sciences & Mathematics --- Mathematics - General --- Statistical matching. --- Statistics --- Formal methods (Computer science) --- Income distribution --- Poverty --- Social sciences --- Data processing. --- Research --- Mathematical models. --- Behavioral sciences --- Human sciences --- Sciences, Social --- Social science --- Social studies --- Destitution --- Distribution of income --- Income inequality --- Inequality of income --- Concatenation, File (Statistics) --- Data fusion (Statistics) --- Data matching (Statistics) --- Data merging (Statistics) --- File concatenation (Statistics) --- Fusion, Data (Statistics) --- Imputation, Mass (Statistics) --- Mass imputation (Statistics) --- Matching, Data (Statistics) --- Matching, Statistical --- Merging, Data (Statistics) --- Microsimulation modeling (Statistics) --- Modeling, Microsimulation (Statistics) --- Applied mathematics. --- Engineering mathematics. --- Computer mathematics. --- Probabilities. --- Wealth --- Basic needs --- Begging --- Poor --- Subsistence economy --- Distribution (Economic theory) --- Disposable income --- System design --- Civilization --- Sampling (Statistics) --- Distribution (Probability theory. --- Computer mathematics --- Discrete mathematics --- Electronic data processing --- Math --- Science --- Distribution functions --- Frequency distribution --- Characteristic functions --- Probabilities --- Engineering --- Engineering analysis --- Mathematical analysis --- Probability --- Statistical inference --- Combinations --- Chance --- Least squares --- Mathematical statistics --- Risk --- Probability Theory.

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Understanding the mechanisms involved in life (e. g. , discovering the biological functionofasetofproteins,inferringtheevolutionofasetofspecies)isbecoming increasinglydependent onprogressmade inmathematics,computer science,and molecular engineering. For the past 30 years, new high-throughput technologies have been developed generating large amounts of data, distributed across many data sources on the Web, with a high degree of semantic heterogeneity and di?erentlevelsofquality. However,onesuchdatasetisnot,byitself,su?cientfor scienti?c discovery. Instead, it must be combined with other data and processed by bioinformatics tools for patterns, similarities, and unusual occurrences to be observed. Both data integration and data mining are thus of paramount importance in life science. DILS 2007 was the fourth in a workshop series that aims at fostering d- cussion, exchange, and innovation in research and development in the areas of data integration and data management for the life sciences. Each previous DILS workshop attracted around 100 researchers from all over the world. This year, the number of submitted papers again increased. The Program Committee - lected 19 papers out of 52 full submissions. The DILS 2007 papers cover a wide spectrum of theoretical and practical issues including scienti?c work?ows, - notation in data integration, mapping and matching techniques, and modeling of life science data. Among the papers, we distinguished 13 papers presenting research on new models, methods, or algorithms and 6 papers presenting imp- mentation of systems or experience with systems in practice. In addition to the presented papers, DILS 2007 featured two keynote talks by Kenneth H. Buetow, National Cancer Institute, and Junhyong Kim, University of Pennsylvania.

Bioinformatics --- Statistical matching --- Computational biology --- Bio-informatique --- Appariement (Statistique) --- Congresses. --- Congrès --- Congresses --- Data integration (Computer science) --- Computing Methodologies --- Medical Informatics Applications --- Biology --- Medical Informatics --- Biological Science Disciplines --- Information Science --- Natural Science Disciplines --- Disciplines and Occupations --- Mathematical Computing --- Computational Biology --- Information Systems --- Computer Simulation --- Engineering & Applied Sciences --- Health & Biological Sciences --- Computer Science --- Biology - General --- Concatenation, File (Statistics) --- Data fusion (Statistics) --- Data matching (Statistics) --- Data merging (Statistics) --- File concatenation (Statistics) --- Fusion, Data (Statistics) --- Imputation, Mass (Statistics) --- Mass imputation (Statistics) --- Matching, Data (Statistics) --- Matching, Statistical --- Merging, Data (Statistics) --- Microsimulation modeling (Statistics) --- Modeling, Microsimulation (Statistics) --- Computer science. --- Health informatics. --- Database management. --- Data mining. --- Bioinformatics. --- Computational biology. --- Computer Science. --- Data Mining and Knowledge Discovery. --- Health Informatics. --- Database Management. --- Information Systems Applications (incl. Internet). --- Computational Biology/Bioinformatics. --- Computer Appl. in Life Sciences. --- Bio-informatics --- Biological informatics --- Information science --- Systems biology --- Algorithmic knowledge discovery --- Factual data analysis --- KDD (Information retrieval) --- Knowledge discovery in data --- Knowledge discovery in databases --- Mining, Data --- Database searching --- Data base management --- Data services (Database management) --- Database management services --- DBMS (Computer science) --- Generalized data management systems --- Services, Database management --- Systems, Database management --- Systems, Generalized database management --- Electronic data processing --- Clinical informatics --- Health informatics --- Medical information science --- Medicine --- Informatics --- Science --- Data processing --- Sampling (Statistics) --- Medical records --- Data processing. --- EHR systems --- EHR technology --- EHRs (Electronic health records) --- Electronic health records --- Electronic medical records --- EMR systems --- EMRs (Electronic medical records) --- Information storage and retrieval systems --- Medical care --- Database management --- Application software. --- Bioinformatics . --- Computational biology . --- Application computer programs --- Application computer software --- Applications software --- Apps (Computer software) --- Computer software