DOCSLIB.ORG

Decision-Making with Multi-Step Expert Advice on the Web

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Decision-Making with Multi-Step Expert Advice on the Web Zur Erlangung des akademischen Grades eines Doktors der Ingenieurwissenschaften (Dr.-Ing.) von der KIT-Fakultät für Wirtschaftswissenschaften des Karlsruher Instituts für Technologie (KIT) genehmigte Dissertation von Patrick Raoul Philipp, M.Sc. Decision-Making with Multi-Step Expert Advice on the Web Patrick Raoul Philipp Tag der mündlichen Prüfung: 13.04.2018 Referent: PD Dr. Achim Rettinger Korreferent: Prof. Dr. Kristian Kersting Karlsruhe, 2018 0 Abstract This thesis deals with solving multi-step tasks by using advice from experts, which are algo- rithms to solve individual steps of such tasks. We contribute with methods for maximizing the number of correct task solutions by selecting and combining experts for individual task instances and methods for automating the process of solving tasks on the Web, where experts are available as Web services. Multi-step tasks frequently occur in Natural Language Processing (NLP) or Computer Vi- sion, and as research progresses an increasing amount of exchangeable experts for the same steps are available on the Web. Service provider platforms such as Algorithmia monetize ex- pert access by making expert services available via their platform and having customers pay for single executions. Such experts can be used to solve diverse tasks, which often consist of multiple steps and thus require pipelines of experts to generate hypotheses. We perceive two distinct problems for solving multi-step tasks with expert services: (1) Given that the task is sufficiently complex, no single pipeline generates correct solutions for all possible task instances. One thus must learn how to construct individual expert pipelines for individual task instances in order to maximize the number of correct solutions, while also taking into account the costs adhered to executing an expert. (2) To automatically solve multi- step tasks with expert services, we need to discover, execute and compose expert pipelines. With mostly textual descriptions of complex functionalities and input parameters, Web au- tomation entails to integrate available expert services and data, interpreting user-specified task goals or efficiently finding correct service configurations. In this thesis, we present solutions to both problems: (1) We enable to learn well-performing expert pipelines assuming available reference data sets (comprising a number of task instances and solutions), where we distinguish between centralized and decentralized decision-making. We formalize the problem as specialization of a Markov Decision Process (MDP), which we refer to as Expert Process (EP) and integrate techniques from Statistical Relational Learning (SRL) or Multiagent coordination. (2) We develop a framework for automatically discovering, executing and composing expert pipelines by exploiting methods developed for the Semantic Web. We lift the representations of experts with structured vocabularies modeled with the Resource Description Framework (RDF) and extend EPs to Semantic Expert Processes (SEPs) to enable the data-driven execution of experts in Web-based architectures. We evaluate our methods in different domains, namely Medical Assistance with tasks in Image Processing and Surgical Phase Recognition, and NLP for textual data on the Web, where we deal with the task of Named Entity Recognition and Disambiguation (NERD). v 0 Publications Text as well as figures in this thesis have partly been published in the following papers: Philipp Gemmeke, Maria Maleshkova, Patrick Philipp, Michael Götz, Christian We- ber, Benedikt Kämpgen, Sascha Zelzer, Klaus Maier-Hein, and Achim Rettinger. Us- ing linked data and web apis for automating the pre-processing of medical images. In Proceedings of the 5th International Workshop on Consuming Linked Data (COLD'14) co-located with the 13th International Semantic Web Conference (ISWC'14), Riva del Garda, Italy, pages 25–36, 2014. [51] Patrick Philipp, Maria Maleshkova, Achim Rettinger, and Darko Katic. A semantic framework for sequential decision making. In Engineering the Web in the Big Data Era - 15th International Conference, ICWE'15, Rotterdam, The Netherlands, pages 392– 409, 2015. doi:10.1007/978-3-319-19890-3_25. [101] Patrick Philipp, Maria Maleshkova, Darko Katic, Christian Weber, Michael Götz, Achim Rettinger, Stefanie Speidel, Benedikt Kämpgen, Marco Nolden, Anna-Laura Wekerle, Rüdiger Dillmann, Hannes Kenngott, Beat P. Müller-Stich, and Rudi Studer. Toward cognitive pipelines of medical assistance algorithms. Int. J. Computer As- sisted Radiology and Surgery, 11(9):1743–1753, 2016. doi: 10.1007/s11548-015-1322- y. [102] Nicolai Schoch, Patrick Philipp, Tobias Weller, Sandy Engelhardt, Mykola Volovyk, Andreas Fetzer, Marco Nolden, Raffaele De Simone, Ivo Wolf, Maria Maleshkova, Achim Rettinger, Rudi Studer, and Vincent Heuveline. Cognitive tools pipeline for assistance of mitral valve surgery. In Proceedings of SPIE, Medical Imaging'16: Image-Guided Procedures, Robotic Interventions and Modeling, pages 9786–9794, 2016. [114] Patrick Philipp and Achim Rettinger. Reinforcement learning for multi-step expert ad- vice. In Proceedings of the 16th Conference on Autonomous Agents and Multiagent Systems, AAMAS'17, São Paulo, Brazil, pages 962–971, 2017. [100] vii Patrick Philipp, Achim Rettinger, and Maria Maleshkova. On automating de- centralized multi-step service combination. In IEEE International Conference on Web Services, ICWS'17, Honolulu, HI, USA, pages 736–743, 2017. doi: 10.1109/ICWS.2017.89. [104] Patrick Philipp, Maria Maleshkova, Achim Rettinger, and Darko Katic. A semantic framework for sequential decision making. J. Web Eng., 16(5&6):471–504, 2017. [103] viii 0 Contents I Foundations1 1 Introduction & Overview3 1.1 Introduction...................................3 1.2 Challenges....................................7 1.2.1 Learning................................7 1.2.2 Automation...............................9 1.3 Research Questions & Hypotheses....................... 10 1.4 Contributions.................................. 12 1.5 Scope of the Thesis............................... 13 1.6 Outline..................................... 15 2 Scenarios 17 2.1 Named Entity Recognition & Disambiguation................. 17 2.1.1 Description............................... 17 2.1.2 Challenges............................... 18 2.2 Surgical Phase Recognition........................... 18 2.2.1 Description............................... 18 2.2.2 Challenges............................... 19 2.3 Tumor Progression Mapping.......................... 19 2.3.1 Description............................... 19 2.3.2 Challenges............................... 20 3 Preliminaries 23 3.1 Multi-Step Tasks................................ 23 3.2 Supervised Machine Learning......................... 27 3.2.1 Model-specific Assumptions...................... 28 3.2.2 Learning Protocols........................... 29 3.2.3 Prediction Settings........................... 30 3.2.4 Selected Learning Scenarios...................... 30 3.2.5 Evaluating Learned Models...................... 33 3.2.6 Learning Combination Functions over Models............ 34 3.3 Decision-Making................................ 34 3.3.1 Prediction with Expert Advice..................... 35 3.3.2 Markov Decision Processes...................... 38 3.3.3 Multiagent Decision Processes..................... 42 3.4 Semantic Web.................................. 44 3.4.1 Resource Description Framework................... 45 ix Contents 3.4.2 RDF Vocabularies........................... 45 3.4.3 Notation3................................ 46 3.4.4 SPARQL................................ 46 3.4.5 Linked Data............................... 47 3.4.6 Services in the Semantic Web..................... 48 4 Related Work 51 4.1 Multi-Step Tasks in Single-Agent Systems................... 51 4.1.1 Learning- & Decision-Theoretic Approaches............. 51 4.1.2 Service Approaches.......................... 53 4.2 Multi-Step Tasks in Multiagent Systems.................... 53 4.2.1 Multiagent Systems for Single-Agent Tasks.............. 53 4.2.2 Multiagent Systems for Multiagent Tasks............... 54 4.3 Multi-Step Tasks on the Web.......................... 55 4.3.1 Semantic Web Service Description Languages............ 55 4.3.2 Decision-Making Frameworks & Applications............ 55 4.3.3 Workflow Systems........................... 56 II Learning 59 5 Learning with Expert Processes 61 5.1 Introduction................................... 61 5.1.1 Challenges............................... 63 5.1.2 Contributions.............................. 64 5.2 Problem Formalization............................. 64 5.3 Meta Dependencies............................... 68 5.3.1 Single Experts............................. 70 5.3.2 Pairwise Intra-Step Experts....................... 71 5.3.3 Pairwise Inter-Step Experts....................... 71 5.4 Online Reinforcement Learning........................ 74 5.4.1 EWH with Meta Dependencies..................... 74 5.4.2 EWH with Incomplete Information.................. 75 5.5 Batch Reinforcement Learning......................... 77 5.5.1 Probabilistic Soft Logic and Hinge-Loss Markov Random Fields... 78 5.5.2 Meta Dependencies with PSL..................... 79 5.6 Discussion.................................... 81 5.7 Summary.................................... 81 6 Learning with Multiagent Expert Processes 83 6.1 Introduction................................... 83 6.1.1 Challenges............................... 85 6.1.2 Contributions.............................. 85 6.2
Load more

Recommended publications
  • Naiad: a Timely Dataflow System
    Naiad: A Timely Dataflow System Derek G. Murray Frank McSherry Rebecca Isaacs Michael Isard Paul Barham Mart´ın Abadi Microsoft Research Silicon Valley {derekmur,mcsherry,risaacs,misard,pbar,abadi}@microsoft.com Abstract User queries Low-latency query are received responses are delivered Naiad is a distributed system for executing data parallel, cyclic dataflow programs. It offers the high throughput Queries are of batch processors, the low latency of stream proces- joined with sors, and the ability to perform iterative and incremental processed data computations. Although existing systems offer some of Complex processing these features, applications that require all three have re- incrementally re- lied on multiple platforms, at the expense of efficiency, Updates to executes to reflect maintainability, and simplicity. Naiad resolves the com- data arrive changed data plexities of combining these features in one framework. A new computational model, timely dataflow, under- Figure 1: A Naiad application that supports real- lies Naiad and captures opportunities for parallelism time queries on continually updated data. The across a wide class of algorithms. This model enriches dashed rectangle represents iterative processing that dataflow computation with timestamps that represent incrementally updates as new data arrive. logical points in the computation and provide the basis for an efficient, lightweight coordination mechanism. requirements: the application performs iterative process- We show that many powerful high-level programming ing on a real-time data stream, and supports interac- models can be built on Naiad’s low-level primitives, en- tive queries on a fresh, consistent view of the results. abling such diverse tasks as streaming data analysis, it- However, no existing system satisfies all three require- erative machine learning, and interactive graph mining.
    [Show full text]
  • ML Cheatsheet Documentation
    ML Cheatsheet Documentation Team Sep 02, 2021 Basics 1 Linear Regression 3 2 Gradient Descent 21 3 Logistic Regression 25 4 Glossary 39 5 Calculus 45 6 Linear Algebra 57 7 Probability 67 8 Statistics 69 9 Notation 71 10 Concepts 75 11 Forwardpropagation 81 12 Backpropagation 91 13 Activation Functions 97 14 Layers 105 15 Loss Functions 117 16 Optimizers 121 17 Regularization 127 18 Architectures 137 19 Classification Algorithms 151 20 Clustering Algorithms 157 i 21 Regression Algorithms 159 22 Reinforcement Learning 161 23 Datasets 165 24 Libraries 181 25 Papers 211 26 Other Content 217 27 Contribute 223 ii ML Cheatsheet Documentation Brief visual explanations of machine learning concepts with diagrams, code examples and links to resources for learning more. Warning: This document is under early stage development. If you find errors, please raise an issue or contribute a better definition! Basics 1 ML Cheatsheet Documentation 2 Basics CHAPTER 1 Linear Regression • Introduction • Simple regression – Making predictions – Cost function – Gradient descent – Training – Model evaluation – Summary • Multivariable regression – Growing complexity – Normalization – Making predictions – Initialize weights – Cost function – Gradient descent – Simplifying with matrices – Bias term – Model evaluation 3 ML Cheatsheet Documentation 1.1 Introduction Linear Regression is a supervised machine learning algorithm where the predicted output is continuous and has a constant slope. It’s used to predict values within a continuous range, (e.g. sales, price) rather than trying to classify them into categories (e.g. cat, dog). There are two main types: Simple regression Simple linear regression uses traditional slope-intercept form, where m and b are the variables our algorithm will try to “learn” to produce the most accurate predictions.
    [Show full text]
  • 1.2 Le Zhang, Microsoft
    Objective • “Taking recommendation technology to the masses” • Helping researchers and developers to quickly select, prototype, demonstrate, and productionize a recommender system • Accelerating enterprise-grade development and deployment of a recommender system into production • Key takeaways of the talk • Systematic overview of the recommendation technology from a pragmatic perspective • Best practices (with example codes) in developing recommender systems • State-of-the-art academic research in recommendation algorithms Outline • Recommendation system in modern business (10min) • Recommendation algorithms and implementations (20min) • End to end example of building a scalable recommender (10min) • Q & A (5min) Recommendation system in modern business “35% of what consumers purchase on Amazon and 75% of what they watch on Netflix come from recommendations algorithms” McKinsey & Co Challenges Limited resource Fragmented solutions Fast-growing area New algorithms sprout There is limited reference every day – not many Packages/tools/modules off- and guidance to build a people have such the-shelf are very recommender system on expertise to implement fragmented, not scalable, scale to support and deploy a and not well compatible with enterprise-grade recommender by using each other scenarios the state-of-the-arts algorithms Microsoft/Recommenders • Microsoft/Recommenders • Collaborative development efforts of Microsoft Cloud & AI data scientists, Microsoft Research researchers, academia researchers, etc. • Github url: https://github.com/Microsoft/Recommenders • Contents • Utilities: modular functions for model creation, data manipulation, evaluation, etc. • Algorithms: SVD, SAR, ALS, NCF, Wide&Deep, xDeepFM, DKN, etc. • Notebooks: HOW-TO examples for end to end recommender building. • Highlights • 3700+ stars on GitHub • Featured in YC Hacker News, O’Reily Data Newsletter, GitHub weekly trending list, etc.
    [Show full text]
  • Alibaba Amazon
    Online Virtual Sponsor Expo Sunday December 6th Amazon | Science Alibaba Neural Magic Facebook Scale AI IBM Sony Apple Quantum Black Benevolent AI Zalando Kuaishou Cruise Ant Group | Alipay Microsoft Wild Me Deep Genomics Netflix Research Google Research CausaLens Hudson River Trading 1 TALKS & PANELS Scikit-learn and Fairness, Tools and Challenges 5 AM PST (1 PM UTC) Speaker: Adrin Jalali Fairness, accountability, and transparency in machine learning have become a major part of the ML discourse. Since these issues have attracted attention from the public, and certain legislations are being put in place regulating the usage of machine learning in certain domains, the industry has been catching up with the topic and a few groups have been developing toolboxes to allow practitioners incorporate fairness constraints into their pipelines and make their models more transparent and accountable. AIF360 and fairlearn are just two examples available in Python. On the machine learning side, scikit-learn has been one of the most commonly used libraries which has been extended by third party libraries such as XGBoost and imbalanced-learn. However, when it comes to incorporating fairness constraints in a usual scikit- learn pipeline, there are challenges and limitations related to the API, which has made developing a scikit-learn compatible fairness focused package challenging and hampering the adoption of these tools in the industry. In this talk, we start with a common classification pipeline, then we assess fairness/bias of the data/outputs using disparate impact ratio as an example metric, and finally mitigate the unfair outputs and search for hyperparameters which give the best accuracy while satisfying fairness constraints.
    [Show full text]
  • Microsoft Recommenders Tools to Accelerate Developing Recommender Systems Scott Graham, Jun Ki Min, Tao Wu {Scott.Graham, Jun.Min, Tao.Wu} @ Microsoft.Com
    Microsoft Recommenders Tools to Accelerate Developing Recommender Systems Scott Graham, Jun Ki Min, Tao Wu {Scott.Graham, Jun.Min, Tao.Wu} @ microsoft.com Microsoft Recommenders repository is an open source collection of python utilities and Jupyter notebooks to help accelerate the process of designing, evaluating, and deploying recommender systems. The repository is actively maintained and constantly being improved and extended. It is publicly available on GitHub and licensed through a permissive MIT License to promote widespread use of the tools. Core components reco_utils • Dataset helpers and splitters • Ranking & rating metrics • Hyperparameter tuning helpers • Operationalization utilities notebooks • Spark ALS • Inhouse: SAR, Vowpal Wabbit, LightGBM, RLRMC, xDeepFM • Deep learning: NCF, Wide-Deep • Open source frameworks: Surprise, FastAI tests • Unit tests • Smoke & Integration tests Workflow Train set Recommender train Environment Data Data split setup load Validation set Hyperparameter tune Operationalize Test set Evaluate • Rating metrics • Content based models • Clone & conda-install on a local / • Chronological split • Ranking metrics • Collaborative-filtering models virtual machine (Windows / Linux / • Stratified split • Classification metrics • Deep-learning models Mac) or Databricks • Non-overlap split • Docker • Azure Machine Learning service (AzureML) • Random split • Tuning libraries (NNI, HyperOpt) • pip install • Azure Kubernetes Service (AKS) • Tuning services (AzureML) • Azure Databricks Example: Movie Recommendation
    [Show full text]
  • The Design and Implementation of Low-Latency Prediction Serving Systems
    The Design and Implementation of Low-Latency Prediction Serving Systems Daniel Crankshaw Electrical Engineering and Computer Sciences University of California at Berkeley Technical Report No. UCB/EECS-2019-171 http://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-171.html December 16, 2019 Copyright © 2019, by the author(s). All rights reserved. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission. Acknowledgement To my parents The Design and Implementation of Low-Latency Prediction Serving Systems by Daniel Crankshaw A dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy in Computer Science in the Graduate Division of the University of California, Berkeley Committee in charge: Professor Joseph Gonzalez, Chair Professor Ion Stoica Professor Fernando Perez Fall 2019 The Design and Implementation of Low-Latency Prediction Serving Systems Copyright 2019 by Daniel Crankshaw 1 Abstract The Design and Implementation of Low-Latency Prediction Serving Systems by Daniel Crankshaw Doctor of Philosophy in Computer Science University of California, Berkeley Professor Joseph Gonzalez, Chair Machine learning is being deployed in a growing number of applications which demand real- time, accurate, and cost-efficient predictions under heavy query load. These applications employ a variety of machine learning frameworks and models, often composing several models within the same application.
    [Show full text]
  • With Honors Final Year Peoject 1
    FACULTY OF INFORMATICS AND COMPUTING BACHELOR OF COMPUTER SCIENCE (COMPUTER NETWORK SECURITY) WITH HONORS FINAL YEAR PEOJECT 1 CSF 35104 3RD YEAR 5TH SEMESTER SESSION 2018/2019 SPAM DETECTION USING MACHINE LEARNING BASED BINARY CLASSIFIER PREPARED BY : NOR SYAIDATUL AMIRAH BINTI ABDUL HAMID MATRIC NUMBER : BTBL16043660 SUPERVISOR : MR. AHMAD FAISAL AMRI BIN ABIDIN@BHARUN CHAPTER 1 INTRODUCTION 1.1 BACKGROUND Today, spam has become a big trouble over the internet. Up to 2017, the statistic shown spam accounted for 55%of all e-mail messages, same as during the previous year [1]. Spam which is also known as unsolicited bulk email has led to the increasing use of email as email provides the perfect ways to send the unwanted advertisement or junk newsgroup posting at no cost for the sender. This chances has been extensively exploited by irresponsible organizations and resulting to clutter the mail boxes of millions of people all around the world. Evolving from a minor to major concern, given the high offensive content of messages, spam is a waste of time. It also consumed a lot of storage space and communication bandwidth. End user is at risk of deleting legitimate mail by mistake. Moreover, spam also impacted the economical which led some countries to adopt legislation. Text classification is used to determine the path of incoming mail/message either into inbox or straight to spam folder. It is the process of assigning categories to text according to its content. It is used to organized, structures and categorize text. It can be done either manually or automatically. Machine learning automatically classifies the text in a much faster way than manual technique.
    [Show full text]
  • Anna Choromanska
    ANNA CHOROMANSKA CONTACT Department of Electrical and Computer Engineering NYU Tandon School of Engineering New York University Room LC266D, 5 Metrotech Center, New York, NY 11201, USA E-mail: ac5455 at nyu dot edu, achoroma at gmail dot com Website: https://engineering.nyu.edu/faculty/anna-choromanska EDUCATION • Columbia University in the City of New York, Department of Electrical Engineering, M.Phil. and Ph.D. (with Presidential Fellowship), 2009-2014 (Advisor: Prof. Tony Jebara, Co-advisor: Prof. Claire Monteleoni) • Warsaw University of Technology, Department of Electronics and Information Technology, MSc with distinctions (double specialization: Electronics and Computer Engineering and Electron- ics and Informatics in Medicine), 2004-2009 (Advisor: Prof. Antoni Grzanka) • Mieczyslaw Karlowicz Music High School in Warsaw, Department of Piano Play, 1998-2004 PROFESSIONAL EXPERIENCE Assistant Professor Department of Electrical and Computer 01.2017 now Engineering, NYU Tandon School of Engineering Affiliated Faculty Member NYU Center for Urban Science 02.2019 now and Progress Affiliated Faculty Member NYU Center for Data Science 01.2017 now ||||||||||||||||||||||||||||||||||||||||||| Research Collaboration NYU - NVIDIA AI Labs (NVAIL) 02.2019 now Working on deep-learning-based AI platforms for autonomous driving. Research Collaboration IBM T. J. Watson Research Center 08.2017 now Working on biologically plausible algorithms for training deep networks. Research Collaboration NVIDIA (New Jersey lab) 05.2016 now Working on autonomous car driving. Post-Doctoral Associate Computer Science Department, 04.2014 12.2016 Courant Institute of Mathematical Sciences, New York University Working on deep learning (advisor: Prof. Yann LeCun). Summer Internship Microsoft Research, New York 06.2013 09.2013 Research Collaboration 09.2013 06.2014 Working on logarithmic time extreme multiclass classification (advisor: Dr John Langford).
    [Show full text]
  • Extreme Multi-Label Classification for Information Retrieval XMLC4IR
    Extreme Multi-label Classification for Information Retrieval XMLC4IR Tutorial at ECIR 2018 Rohit Babbar1 and Krzysztof Dembczy´nski2 1Aalto University, Finland, 2Pozna´nUniversity of Technology, Poland European Conference on Information Retrieval Grenoble, France, March 26, 2018 • Rohit Babbar: I Affiliation: University Aalto I Previous affiliations: Max-Planck Institute for Intelligent Systems, Universit´eGrenoble Alpes I Main interests: extreme classification, multi-label classification, multi-class classification, text classification 1 / 94 • Krzysztof Dembczy´nski: I Affiliation: Poznan University of Technology I Previous affiliations: Marburg University I Main interests: extreme multi-machine learning, multi-label classification, label tree algorithms, learning theory 2 / 94 Agenda 1 Extreme classification: applications and challenges 2 Algorithms I Label embeddings I Smart 1-vs-All approaches I Tree-based methods I Label filtering/maximum inner product search 3 Live demonstration Webpage: http://www.cs.put.poznan.pl/ kdembczynski/xmlc-tutorial-ecir-2018/ 3 / 94 Agenda 1 Extreme classification: applications and challenges 2 Algorithms I Label embeddings I Smart 1-vs-All approaches I Tree-based methods I Label filtering/maximum inner product search 3 Live demonstration Webpage: http://www.cs.put.poznan.pl/ kdembczynski/xmlc-tutorial-ecir-2018/ 3 / 94 Extreme multi-label classification is a problem of labeling an item with a small set of tags out of an extremely large number of potential tags. 4 / 94 Geoff Hinton Andrew Ng Yann LeCun Yoshua Bengio 5 /
    [Show full text]
  • Analytical and Other Software on the Secure Research Environment
    Analytical and other software on the Secure Research Environment The Research Environment runs with the operating system: Microsoft Windows Server 2016 DataCenter edition. The environment is based on the Microsoft Data Science virtual machine template and includes the following software: • R Version 4.0.2 (2020-06-22), as part of Microsoft R Open • R Studio Desktop 1.3.1093 working with R 4.0.2 • Anaconda 3, including an environment for Python 3.8.5 • Python, 3.8.5 as part of the Anaconda base environment • Jupyter Notebook, as part of the Anaconda3 environment • Microsoft Office 2016 Standard edition, including Word, Excel, PowerPoint, and OneNote (Access not included) • JuliaPro 0.5.1.1 and the Juno IDE for Julia • PyCharm Community Edition, 2020.3 • PLINK • JAGS • WinBUGS • OpenBUGS • stan and rstan • Apache Spark 2.2.0 • SparkML and pySpark • Apache Drill 1.11.0 • MAPR Drill driver • VIM 8.0.606 • TensorFlow • MXNet, MXNet Model Server • Microsoft Cognitive Toolkit (CNTK) • Weka • Vowpal Wabbit • xgboost • Team Data Science Process (TDSP) Utilities • VOTT (Visual Object Tagging Tool) 1.6.11 • Microsoft Machine Learning Server • PowerBI • Docker version 10.03.5, build 2ee0c57608 • SQL Server Developer Edition (2017), including Management Studio and SQL Server Integration Services (SSIS) • Visual Studio Code 1.17.1 • Nodejs • 7-zip • Evince PDF Viewer • Acrobat Reader • Microsoft Photo Viewer • PowerShell 6 March 2021 Version 1.2 And in the Premium research environments: • STATA 16.1 • SAS 9.4, m4 (academic license) Users also have the ability to bring in additional software if the software was specified in the data request, the software runs in the operating system described above, and the user can provide Vivli with any necessary licensing keys.
    [Show full text]
  • Azureml: Anatomy of a Machine Learning Service
    JMLR: Workshop and Conference Proceedings 50:1–13, 2016 PAPIs 2015 AzureML: Anatomy of a machine learning service AzureML Team, Microsoft ∗ 1 Memorial Drive, Cambridge, MA 02142 Editor: Louis Dorard, Mark D. Reid and Francisco J. Martin Abstract In this paper, we describe AzureML, a web service that provides a model authoring en- vironment where data scientists can create machine learning models and publish them easily (http://www.azure.com/ml). In addition, AzureML provides several distinguishing features. These include: (a) collaboration, (b) versioning, (c) visual workflows, (d) exter- nal language support, (e) push-button operationalization, (f) monetization and (g) service tiers. We outline the system overview, design principles and lessons learned in building such a system. Keywords: Machine learning, predictive API 1. Introduction With the rise of big-data, machine learning has moved from being an academic interest to providing competitive advantage to data driven companies. Previously, building such predic- tive systems required deep expertise in machine learning as well as scalable software engineer- ing. Recent trends have democratized machine learning and made building predictive appli- cations far easier. Firstly, the emergence of machine learning libraries such as scikit-learn Pe- dregosa et al.(2011), WEKA Holmes et al.(1994), Vowpal Wabbit Langford et al.(2007) etc. and open source languages like R R Development Core Team(2011) and Julia Bezanson et al. (2012) have made building machine learning models easier even without machine learning expertise. Secondly, scalable data frameworks Zaharia et al.(2010); Low et al.(2010) have provided necessary infrastructure for handling large amounts of data. Operationalizing mod- els built this way still remains a challenge.
    [Show full text]
  • Package 'Rvowpalwabbit'
    Package ‘RVowpalWabbit’ August 7, 2020 Title R Interface to the Vowpal Wabbit Version 0.0.15 Date 2020-08-03 Maintainer Dirk Eddelbuettel <[email protected]> Author Dirk Eddelbuettel Description The 'Vowpal Wabbit' project is a fast out-of-core learning system sponsored by Microsoft Research (having started at Yahoo! Research) and written by John Langford along with a number of contributors. This R package does not include the distributed computing implementation of the cluster/ directory of the upstream sources. Use of the software as a network service is also not directly supported as the aim is a simpler direct call from R for validation and comparison. Note that this package contains an embedded older version of 'Vowpal Wabbit'. The package 'rvw' at the GitHub repo <https://github.com/eddelbuettel/rvw> can provide an alternative using an external 'Vowpal Wabbit' library installation. Depends R (>= 2.12.0) Imports Rcpp LinkingTo Rcpp SystemRequirements The Boost 'program_options' library <https://boost.org> is required. OS_type unix StagedInstall no License GPL (>= 2) URL https://vowpalwabbit.org/ BugReports https://github.com/eddelbuettel/rvowpalwabbit/issues NeedsCompilation yes Repository CRAN Date/Publication 2020-08-07 08:40:15 UTC 1 2 vw R topics documented: vw..............................................2 Index 4 vw Run the Vowpal Wabbit fast out-of-core learner Description The vw function applies the Vowpal Wabbit on-line learner to a given data set and model. Vowpal Wabbit is a project sponsored by Yahoo! Research and led by John Langford. At present, this package provides a simple yet crude interface. Usage vw(args, quiet=TRUE) Arguments args A character vector containing the same arguments one would use on the command- line with the standalone vw binary.
    [Show full text]