Amidst demo (BNAIC 2015)
0 likes118 views
The document describes the AMIDST Toolbox, an open source Java library for analyzing massive data streams using probabilistic graphical models. The toolbox uses Bayesian networks and inference algorithms like variational message passing to model and learn from continuous data streams in a scalable way. It was developed as part of an EU research project to handle real-world complex streaming data. The document outlines the toolbox's capabilities and provides an example live demo of using it to model concept drift in financial data by tracking changes over time with a hidden variable in a naive Bayes model.
![Querying the model
§ Parallel Monte Carlo Inference [Salmeron et al. CAEPIA 2015]
§ Exploit Multi-Core (powered by Java 8)
BNAIC 2015 November 5-6, 2015](https://image.slidesharecdn.com/amidst-demo-160921115328/85/Amidst-demo-BNAIC-2015-20-320.jpg)
![Querying the model
§ Parallel Monte Carlo Inference [Salmeron et al. CAEPIA 2015]
§ Exploit Multi-Core (powered by Java 8)
§ Variational Message Passing [Winn et al. JMLR 2004]
§ Deterministic approximation
BNAIC 2015 November 5-6, 2015](https://image.slidesharecdn.com/amidst-demo-160921115328/85/Amidst-demo-BNAIC-2015-21-320.jpg)
![Learning from data streams
§ Parallel Streaming Variational Bayes[Broderick et al. NIPS 2013]
§ Powered by Variational Message Passing.
§ Multi-core processing (using Java 8).
BNAIC 2015 November 5-6, 2015](https://image.slidesharecdn.com/amidst-demo-160921115328/85/Amidst-demo-BNAIC-2015-24-320.jpg)
Amidst demo (BNAIC 2015)
- 1. 1BNAIC 2015 November 5-6, 2015 AMIDST Toolbox A Java library for Analysis of MassIve Data Streams using Probabilistic Graphical Models FP7 European research project http://amidst.eu Anders L. Madsen, Andres R. Masegosa, Ana M. Martinez, Hanen Borchani, Thomas D. Nielsen, Helge Langseth, Antonio Salmeron, Dario Ramos-Lopez.
- 2. Outline 1. Overview of AMIDST Toolbox o Why data streams are important? o Why PGMs for analyzing data streams? o Scalable Inference (and learning) o Roadmap for coming releases 2. Live Demo: Modeling concept drift in financial data. o Handling data streams. o Defining Bayesian networks with hidden variables. o Inference and Learning Bayesian networks. BNAIC 2015 November 5-6, 2015
- 3. Scope Part I
- 4. Data Streams everywhere • Unbounded flows of data are generated daily: • Social Networks • Network Monitoring • Financial/Banking industry • …. BNAIC 2015 November 5-6, 2015
- 5. Data Stream Processing • Processing data streams is challenging: – Do not fit in main memory – Continuous Model updating – Continuous Model Inference – Concept Drift BNAIC 2015 November 5-6, 2015
- 6. Processing Massive Data Streams • Everything has to scale: • Scalable Computing infrastructure • Scalable Models/Inference/Learning BNAIC 2015 November 5-6, 2015
- 7. AMIDST Toolbox • Scalable framework for data stream processing. • Based on Probabilistic Graphical Models. • Unique project for data stream mining using PGMs. • Open source project (Apache Software License 2.0). BNAIC 2015 November 5-6, 2015
- 8. AMIDST EU Project 8 § This toolbox aims to deal with real, complex and massive data streams. § Applied to real use-cases of AMIDST’s industrial partners. BNAIC 2015 November 5-6, 2015
- 9. Toolbox Web Page http://amidst.github.io/toolbox/ BNAIC 2015 November 5-6, 2015
- 11. Why Graphical Models? § Let’s look at the following simple example: § Stream of sensor measurements about temperature and smoke presence in a given geographical area. § Monitor the stream to detect the presence of a fire (event detection problem) ? BNAIC 2015 November 5-6, 2015
- 12. § Cast the problem as an anomaly detection problem (outliers). § Streaming K-Means (widely used in industry). Why Graphical Models? Anomaly BNAIC 2015 November 5-6, 2015
- 13. Why Graphical Models for analyzing Data Streams? § Many data streams models are black box models: § Pros: § No need to understand the problem. § Cons: § Many hyper-parameters to tune. § Blackbox models can rarely explain what they learned. Stream Blackbox Model Predictions BNAIC 2015 November 5-6, 2015
- 14. § Bayesian Networks: § Openbox models § Encode prior knowledge. § Continuous and discrete variables (CLG networks). § Example: Why Graphical Models? Fire Temp Smoke T1 T2 T3 S1 p(Fire=true|t1,t2,t3,s1) BNAIC 2015 November 5-6, 2015
- 15. Why Graphical Models? Stream Predictions Openbox Models BNAIC 2015 November 5-6, 2015
- 16. Why Graphical Models? Stream Predictions Openbox Models Blackbox Inference Engine (multi-core parallelization) BNAIC 2015 November 5-6, 2015
- 18. Inference Engine § Querying the model § p(Fire=true|t1,t2,t3,s1,season) § E(Temperature|smoke=true). BNAIC 2015 November 5-6, 2015
- 19. Inference Engine § Querying the model § p(Fire=true|t1,t2,t3,s1,season) § E(Temperature|smoke=true) § Learning from data (using a Bayesian approach): § Bayesian framework naturally deals with data streams. § Prior is updated in the light of new data. p(✓|d1, . . . , dn, dn+1) / p(dn+1|✓)p(✓|d1, . . . , dn) BNAIC 2015 November 5-6, 2015
- 20. Querying the model § Parallel Monte Carlo Inference [Salmeron et al. CAEPIA 2015] § Exploit Multi-Core (powered by Java 8) BNAIC 2015 November 5-6, 2015
- 21. Querying the model § Parallel Monte Carlo Inference [Salmeron et al. CAEPIA 2015] § Exploit Multi-Core (powered by Java 8) § Variational Message Passing [Winn et al. JMLR 2004] § Deterministic approximation BNAIC 2015 November 5-6, 2015
- 22. Learning from data streams § Bayesian approach: § Learning as an inference problem. § Powered by VMP. ✓ Z x i = 1 . . . N ↵ BNAIC 2015 November 5-6, 2015
- 23. Learning from data streams § Bayesian approach: § Learning as an inference problem. § Powered by VMP. § Plateau notation!! BNAIC 2015 November 5-6, 2015
- 24. Learning from data streams § Parallel Streaming Variational Bayes[Broderick et al. NIPS 2013] § Powered by Variational Message Passing. § Multi-core processing (using Java 8). BNAIC 2015 November 5-6, 2015
- 25. Links to other open software § MoaLink § MOA is a state-of-the-art tool for data stream mining. § Using AMIDST models within MOA GUI! § Great for evaluation & comparison. BNAIC 2015 November 5-6, 2015
- 26. Links to other open software § HuginLink § Hugin is a commercial software for PGMs and influence diagrams. § Models conversion. § Hugin inference engine can be used within AMIDST. 26BNAIC 2015 November 5-6, 2015
- 28. Dynamic Bayesian Networks (release 1.1) § Encode temporal knowledge § Naturally fits with data streams Fire(t) Temp(t) Smoke(t) T1(t) T2(t) T3(t) S1(t) Fire(t-1) Temp(t-1) BNAIC 2015 November 5-6, 2015
- 29. Distributed Stream Processing (release 1.1) § RLink § Invoke AMIDST Inference engine within R. § Preliminary functionality recently presented. 29BNAIC 2015 November 5-6, 2015
- 30. Distributed Stream Processing (release 2.0) § FlinkLink § Apache Flink: Open source platform for distributed stream processing. § Handling Massive Data Streams. 30BNAIC 2015 November 5-6, 2015
- 31. Open Source project § We’re open to your contributions!! ;) 31BNAIC 2015 November 5-6, 2015
- 32. Hosted on Github § Download: :> git clone https://github.com/amidst/toolbox.git § Compile: :> ./compile.sh § Run: :> ./run.sh <class-name> BNAIC 2015 November 5-6, 2015
- 33. Please “star” our project! (if you like it) 33BNAIC 2015 November 5-6, 2015
- 34. Any questions before the live demo ? 34
- 36. Demo Code Available in Github 36 eu.amidst.bnaic2015.examples.BCC BNAIC 2015 November 5-6, 2015
- 37. Financial Data § Provided by BCC (spanish regional bank). § Consist of monthly aggregated information § Active clients between 18 and 65 years old. § Data between April 2007 and March 2014. § 11 variables § Income, total credit, expenses, etc. § Each client is classified as: § defaulter/non-defaulter in following 12 months. 37BNAIC 2015 November 5-6, 2015
- 38. Financial Data § Hypothesis: § Does spanish financial crisis impact on bank customers? § Look at the evolution of regional unemployment rate. 38BNAIC 2015 November 5-6, 2015
- 39. Data Preprocessing/Visualization § Visualize the evolution of the monthly aggregated data: § Data does not fit in main memory! 39BNAIC 2015 November 5-6, 2015
- 40. Model Building § We use a simple Naïve Bayes model: § With a global hidden variable to track concept drift. 40 D A1 A2 A11… H BNAIC 2015 November 5-6, 2015
- 41. Model Building § We also use Plateau notation § “H” is designed to capture concept drift 41 D A1 A2 A11… HtHt-1 i=1…M ✓ BNAIC 2015 November 5-6, 2015
- 42. Tracking concept drift 42BNAIC 2015 November 5-6, 2015
- 43. Tracking concept drift 43BNAIC 2015 November 5-6, 2015
- 44. References § Masegosa et al. AMIDST: Analysis of Massive Data Streams using Probabilistic Graphical Models. Submitted to JMLR. 2015. § Borchani et al. Modeling Concept Drift: A Probabilistic Graphical Model Based Approach. IDA 2015. § Masegosa et al. Probabilistic graphical models on multi-core CPUs using Java 8. Submitted to IEEE Computational Intelligence Magazine, Special Issue on Computational Intelligence Software. 2015. § Salmeron et al. Parallel importance sampling in conditional linear Gaussian networks. In Proceedings of the Conferencia de la Asociacion Española para la Inteligencia Artificial, volume in press, 2015. § Winn et al. Variational message passing. Journal of Machine Learning Research, 6:661–694, 2005. § Broderick et al. Streaming variational Bayes. In Advances in Neural Information Processing Systems, pages 1727–1735, 2013. 44BNAIC 2015 November 5-6, 2015
- 45. Any questions ? 45 http://amidst.github.io/toolbox/ BNAIC 2015 November 5-6, 2015
- 46. Open Source project § We’re open to your contributions!! ;) 46BNAIC 2015 November 5-6, 2015