Peter Sarlin

Peter Sarlin PhD icon

PhD in Applied Machine Learning

Professor of Practice at Hanken

Silo.AI Peter Sarlin

Co-founder & CEO Silo.AI

Biography

Peter Sarlin is the CEO and Co-Founder of Silo.AI, the largest private artificial intelligence lab in the Nordics. He is also a Professor of Practice specializing in machine learning and artificial intelligence at Hanken School of Economics (Helsinki, Finland).

As a founder of RiskLab Finland at Arcada, Peter is a co-organizer of the Annual RiskLab/BoF/ESRB Systemic Risk Analytics Conference and Europe’s 1st Fintech Master’s program. He is also a research associate with the Systemic Risk Center at London School of Economics, Imperial College London, IWH Halle Institute for Economic Research and the Financial Innovation Lab at University of Cape Town, as well as a board member of the IEEE Analytics and Risk Technical Committee and the IEEE Computational Finance and Economics Technical Committee. Moreover, he is an Associate Editor of Journal of Network Theory in Finance and Intelligent Systems in Accounting, Finance & Management. Peter completed his PhD in 2013 at Turku Centre for Computer Science, and has also studied at London School of Economics, Stockholm School of Economics and Stockholm University.

Peter has built solutions as an external consultant with the European Central Bank, International Monetary Fund, Bank of Finland, Deutsche Bundesbank, De Nederlandsche Bank, Bank of Indonesia and Banco de la República of Colombia, as well as several other private organizations. He is also a founder of Infolytika Ventures and Almax Analytics.  Peter’s book Mapping Financial Stability was published by Springer in May 2014 and his current research interests include machine learning, natural-language processing, complex systems and visual analytics.

Publications

Work in progress

  • Using Artificial Intelligence to Create Value in Insurance, with Riikkinen M, Saarijärvi H, Lähteenmäki I. Submitted.
  • A framework for early-warning modeling. Submitted.
  • RiskRank to predict systemic banking crises with common exposures, with Giudici P, Spelta A, Björk K-M. Submitted.
  • Weighted crisis signals: A country versus a global observer.
  • State of the art in crisis prediction: A literature review and a tool for modeling, with Holopainen M.

Working papers

  • Cerchiello P, Nicola G, Rönnqvist S, Sarlin P, 2016. Deep Learning Bank Distress from News and Numerical Financial Data. [UNIPV] Submitted.
  • Forss T, Sarlin P, 2017. News-sentiment networks as a company risk indicator. [arXiv] Submitted.
  • Sarlin P, von Schweinitz G, 2015. Optimizing policymakers’ loss functions in crisis prediction: before, within or after? [ECB] [IWH] Submitted
  • Rancan M, Sarlin P, Peltonen T, 2015. Interconnectedness of the banking sector as a vulnerability to crises. [ECB] [SSRN] Submitted.

Selected publications

2017
1.
Holopainen, M. & Sarlin, P. Toward robust early-warning models: A horse race, ensembles and model uncertainty. Quantitative Finance forthcoming (2017). http://doi.org/10.1080/14697688.2017.1357972
2.
Rönnqvist, S. & Sarlin, P. Bank distress in the news: Describing events through deep learning. Neurocomputing 264, 57–70 (2017). http://doi.org/10.1016/j.neucom.2016.12.110
3.
Mezei, J. & Sarlin, P. RiskRank: Measuring Interconnected Risk. Economic Modeling forthcoming (2017). http://doi.org/10.1016/j.econmod.2017.04.016
4.
Giudici, P., Sarlin, P. & Spelta, A. The multivariate nature of systemic risk: Direct and common exposures. Journal of Banking & Finance forthcoming (2017). http://doi.org/10.1016/j.jbankfin.2017.05.010
5.
Constantin, A., Peltonen, T. & Sarlin, P. Network linkages to predict bank distress. Journal of Financial Stability forthcoming (2017). http://doi.org/10.1016/j.jfs.2016.10.011
6.
Mezei, J. & Sarlin, P. Introduction to Machine Learning and Network Analytics in Finance Minitrack. in Proceedings of the 2017 Hawaii International Conference on System Sciences (HICSS) (IEEE Press, 2017).
7.
Mezei, J. & Sarlin, P. Possibilistic Clustering for Crisis Prediction: Systemic Risk States and Membership Degrees. in Proceedings of the 2017 Hawaii International Conference on System Sciences (HICSS) (IEEE Press, 2017).
2016
8.
Mezei, J. & Sarlin, P. Aggregating expert knowledge for the measurement of systemic risk. Decision Support Systems 88, 38–50 (2016). http://doi.org/10.1016/j.dss.2016.05.007
9.
Forss, T. & Sarlin, P. From News to Company Networks: Co-occurrence, sentiment, and information centrality. in Proceedings of the 2016 IEEE Conference on Computational Intelligence for Financial Engineering & Economics (CIFEr) (IEEE Press, 2016). http://doi.org/10.1109/SSCI.2016.7850022
10.
Sarlin, P. Editorial on Computational Tools for Systemic Risk Identification and Assessment. Intelligent Systems in Accounting, Finance and Management 23(1–2), 1–2 (2016). http://doi.org/10.1002/isaf.1389
11.
Mezei, J. & Sarlin, P. On interval-valued possibilistic clustering for a generalized objective function. in Proceedings of the 2016 IEEE World Congress on Computational Intelligence (IEEE WCCI) (IEEE Press, 2016). http://doi.org/10.1109/FUZZ-IEEE.2016.7737773
12.
Mezei, J. & Sarlin, P. On a generalized objective function for possibilistic fuzzy clustering. in Proceedings of the 2016 Conference on Information Processing and Management of Uncertainty (IPMU) (Springer-Verlag, 2016). http://doi.org/10.1007/978-3-319-40596-4_59
13.
Oet, P., Gramlich, D. & Sarlin, P. Evaluating measures of adverse financial conditions. Journal of Financial Stability 27, 234–249 (2016). http://doi.org/10.1016/j.jfs.2016.06.008
14.
Sarlin, P. Macroprudential oversight, risk communication and visualization. Journal of Financial Stability 27,  160–179 (2016). http://doi.org/10.1016/j.jfs.2015.12.005
15.
Sarlin, P. Visual Macroprudential Surveillance of Banks. Intelligent Systems in Accounting, Finance and Management 23(4), 257–264 (2016). http://doi.org/10.1002/isaf.1391
16.
Sarlin, P. & Peltonen, T. Introduction to Systemic Risk Analytics Minitrack. in Proceedings of the 2016 Hawaii International Conference on System Sciences (HICSS) (IEEE Press, 2016). http://doi.org/10.1109/HICSS.2016.221
17.
Kouontchou, P. et al. A R-SOM Analysis of the Link between Financial Market Conditions and a Systemic Risk Index based on ICA-factors of Systemic Risk Measures. in Proceedings of the 2016 Hawaii International Conference on System Sciences (HICSS) (IEEE Press, 2016). http://doi.org/10.1109/HICSS.2016.222
18.
Ramsay, B. & Sarlin, P. Ending over-lending: Assessing systemic risk with debt to cash flow. International Journal of Finance & Economics 21(1), 36–57 (2016). http://doi.org/10.1002/ijfe.1520
2015
19.
Rönnqvist, S. & Sarlin, P. Detect & Describe: Deep Learning of Bank Stress in the News. in Proceedings of the 2015 IEEE Conference on Computational Intelligence for Financial Engineering & Economics (CIFEr) (IEEE Press, 2015). http://doi.org/10.1109/SSCI.2015.131
20.
Laina, P., Nyholm, J. & Sarlin, P. Leading indicators of systemic banking crises: Finland in a panel of EU countries. Review of Financial Economics 24, 18–35 (2015). http://doi.org/10.1016/j.rfe.2014.12.002
21.
Holopainen, M. & Sarlin, P. CrisisModeler: A Tool for Exploring Crisis Predictions. in Proceedings of the 2015 IEEE Conference on Computational Intelligence for Financial Engineering & Economics (CIFEr) (IEEE Press, 2015). http://doi.org/10.1109/SSCI.2015.135
22.
Nikou, S., Mezei, J. & Sarlin, P. A process view to evaluate and understand preference elicitation. Journal of Multi-Criteria Decision Analysis 22(5-6), 305–329 (2015). http://doi.org/10.1002/mcda.1544
23.
Rönnqvist, S. & Sarlin, P. Bank Networks from Text: Interrelations, Centrality and Determinants. Quantitative Finance 15(10), 1619–1635 (2015). http://doi.org/10.1080/14697688.2015.1071076
24.
Sarlin, P. & Nyman, H.J. The process of macropudential oversight in Europe. Global Policy 6(4), 389–407 (2015). http://doi.org/10.1111/1758-5899.12255
25.
Sarlin, P. Automated and Weighted Self-Organizing Time Maps. Knowledge and Information Systems 44(2), 493–505 (2015). http://doi.org/10.1007/s10115-014-0762-y
26.
Sarlin, P., Nikou, S., Mezei, J. & Bouwman, H. Visual Conjoint Analysis (VCA): A topology of preferences in multi-attribute decision making. Quality & Quantity 49(1), 385–405 (2015). http://doi.org/10.1007/s11135-014-9992-z.
27.
Sarlin, P. Data and Dimension Reduction for Visual Financial Performance Analysis. Information Visualization 14(2), 148–167 (2015). http://doi.org/10.1177/1473871613504102
28.
Clark, S., Sarlin, P., Sharma, A. & Sisson, S. Increasing dependence on foreign water resources? An assessment of trends in global virtual water flows using a self-organizing time map. Ecological Informatics 26(2), 192–202 (2015). http://doi.org/10.1016/j.ecoinf.2014.05.012
2014
29.
Sarlin, P. Mapping Financial Stability. in Computational Risk Management Series (Springer-Verlag, 2014). http://doi.org/10.1007/978-3-642-54956-4
30.
Sarlin, P. & Nyman, H. J. From Bits to Atoms: 3D Printing in the Context of Supply Chain Strategies. in Proceedings of the 47th Hawaii International Conference on System Sciences (HICSS) (IEEE Press, 2014). http://doi.org/10.1109/HICSS.2014.518
31.
Rönnqvist, S. & Sarlin, P. From Text to Bank Interrelation Maps. in Proceedings of the 2014 IEEE Conference on Computational Intelligence for Financial Engineering & Economics (CIFEr) (IEEE Press, 2014). http://doi.org/10.1109/CIFEr.2014.6924053
32.
Betz, F., Oprica, S., Peltonen, T. A. & Sarlin, P. Predicting Distress in European Banks. Journal of Banking & Finance 45, 225–241 (2014). http://doi.org/10.1016/j.jbankfin.2013.11.041
33.
Yao, Z., Sarlin, P., Eklund, T. & Back, B. Combining Visual Customer Segmentation and Response Modeling. Neural Computing & Applications 25, 123–134 (2014). http://doi.org/10.1007/s00521-013-1454-3
34.
Sarlin, P. A Weighted SOM for classifying data with instance-varying importance. International Journal of Machine Learning and Cybernetics 5(1), 101–110 (2014). http://doi.org/10.1007/s13042-013-0175-3
35.
Sarlin, P. On biologically inspired predictions of the global financial crisis. Neural Computing & Applications 24(3–4), 663–673 (2014). http://doi.org/10.1007/s00521-012-1281-y
2013
36.
Holmbom, A., Sarlin, P., Yao, Z., Eklund, T. & Back, B. Visual Data-Driven Profiling of Green Consumers. in Proceedings of the International Conference on Information Visualization (iV 13) (IEEE Press, 2013). http://doi.org/10.1109/IV.2013.37
37.
Sarlin, P. A Self-Organizing Time Map for time-to-event data. in Proceedings of the IEEE Symposium on Computational Intelligence and Data Mining (CIDM) (IEEE Press, 2013). http://doi.org/10.1109/CIDM.2013.6597241
38.
Holmbom, A., Rönnqvist, S., Sarlin, P., Eklund, T. & Back, B. Green vs. non-green customer behavior: A Self-Organizing Time Map over greenness. in Proceedings of the 13th IEEE International Conference on Data Mining Workshops (ICDMW’13) (IEEE Press, 2013). http://doi.org/10.1109/ICDMW.2013.103
39.
Sarlin, P. & Peltonen, T. Mapping the State of Financial Stability. Journal of International Financial Markets, Institutions & Money 26, 46–76 (2013). http://doi.org/10.1016/j.intfin.2013.05.002
40.
Sarlin, P. & Rönnqvist, S. Cluster coloring of the Self-Organizing Map: An information visualization perspective. in Proceedings of the International Conference on Information Visualization (iV 13) (IEEE Press, 2013). http://doi.org/10.1109/IV.2013.72
41.
Sarlin, P. & Yao, Z. Clustering of the Self-Organizing Time Map. Neurocomputing 121, 317–327 (2013). http://doi.org/10.1016/j.neucom.2013.04.007
42.
Sarlin, P. Decomposing the Global Financial Crisis: A Self-Organizing Time Map. Pattern Recognition Letters 34, 1701–1709 (2013). http://doi.org/10.1016/j.patrec.2013.03.017
43.
Sarlin, P. Exploiting the Self-Organizing Financial Stability Map. Engineering Applications of Artificial Intelligence 26(5–6), 1532–1539 (2013). http://doi.org/10.1016/j.engappai.2013.01.002
44.
Sarlin, P. On policymakers’ loss functions and the evaluation of early warning systems. Economics Letters 119(1), 1–7 (2013). http://doi.org/10.1016/j.econlet.2012.12.030
45.
Sarlin, P. Self-Organizing Time Map: An Abstraction of Temporal Multivariate Patterns. Neurocomputing 99(1), 496–508 (2013). http://doi.org/10.1016/j.neucom.2012.07.011
46.
Sarlin, P. & Eklund, T. Financial Performance Analysis of European Banks Using a Fuzzified Self-Organizing Map. International Journal of Knowledge-Based and Intelligent Engineering Systems 17(3), 223–234 (2013). http://doi.org/10.3233/KES-130261
2012
47.
Sarlin, P., Yao, Z. & Eklund, T. Probabilistic Modeling of State Transitions on the Self-Organizing Map: Some Temporal Financial Applications. in Proceedings of the 45th Hawaii International Conference on System Sciences (HICSS) (IEEE Press, 2012). http://doi.org/10.1109/HICSS.2012.486
48.
Sarlin, P. A Weighted SOM for Classifying Data with Instance-Varying Importance. in Proceedings of the IEEE 12th International Conference on Data Mining Workshops (IEEE Press, 2012). http://doi.org/10.1109/ICDMW.2012.18
49.
Sarlin, P. Chance Discovery with Self-Organizing Maps: Discovering Imbalances in Financial Networks. in Advances in Chance Discovery (eds. Ohsawa, Y. & Abe, A.) 49–61 (Springer-Verlag, 2012). http://doi.org/10.1007/978-3-642-30114-8_4
50.
Sarlin, P., Yao, Z. & Eklund, T. A Framework for State Transitions on The Self-Organizing Map: Some Temporal Financial Applications. Intelligent Systems in Accounting, Finance and Management 19(1), 189–203 (2012). http://doi.org/10.1002/isaf.1328
2012
51.
Sarlin, P. Visual tracking of the millennium development goals with a fuzzified self-organizing neural network. International Journal of Machine Learning and Cybernetics 3(3), 233–245 (2012). http://doi.org/10.1007/s13042-011-0057-5
52.
Yao, Z., Sarlin, P., Eklund, T. & Back, B. Temporal Customer Segmentation Using the Self-Organizing Time Map. in Proceedings of the International Conference on Information Visualisation (iV 12) 234–240 (IEEE Press, 2012). http://doi.org/10.1109/IV.2012.47
2011
53.
Sarlin, P. Visual tracking of the Millennium Development Goals with a Self-organizing neural network. in Proceedings of the IEEE Symposium on Computational Intelligence and Data Mining (CIDM) (IEEE Press, 2011). http://doi.org/10.1109/CIDM.2011.5949433
54.
Sarlin, P. Clustering the Changing Nature of Currency Crises in Emerging Markets: An Exploration with Self-Organising Maps. International Journal of Computational Economics and Econometrics 2(1), 24–46 (2011). http://doi.org/10.1504/IJCEE.2011.040575
55.
Sarlin, P. Sovereign debt monitor: A visual Self-organizing maps approach. in Proceedings of IEEE Symposium on Computational Intelligence for Financial Engineering and Economics (CIFEr 11) 1–8 (IEEE Press, 2011). http://doi.org/978-1-4244-9933-5
56.
Sarlin, P. & Eklund, T. Fuzzy Clustering of the Self-Organizing Map: Some Applications on Financial Time Series. in Proceedings of the Workshop on Self-Organizing Maps (WSOM 11) 40–50 (Springer-Verlag, 2011). http://doi.org/10.1007/978-3-642-21566-7_4
57.
Sarlin, P. & Marghescu, D. Neuro-Genetic Predictions of Currency crises. Intelligent Systems in Accounting, Finance and Management 18(4), 145–160 (2011). http://doi.org/10.1002/isaf.328
58.
Sarlin, P. & Marghescu, D. Visual predictions of currency crises using self-organizing maps. Intelligent Systems in Accounting, Finance and Management18(1), 15–38 (2011). http://doi.org/10.1002/isaf.321
59.
Sarlin, P. Evaluating a Self-Organizing Map for Clustering and Visualizing Optimum Currency Area Criteria. Economics Bulletin 31, 1483–1495 (2011).
2010
60.
Sarlin, P. Visual Predictions of Currency Crises Using Self-Organizing Maps. in Proceedings of the IEEE International Conference on Data Mining Workshops (IEEE Press, 2010). http://doi.org/10.1109/ICDMW.2010.55
61.
Marghescu, D., Sarlin, P. & Liu, S. Early-warning analysis for currency crises in emerging markets: A revisit with fuzzy clustering. Intelligent Systems in Accounting, Finance & Management 17(3–4), 143–165 (2010). http://doi.org/10.1002/isaf.317
62.
Sarlin, P. Visual monitoring of financial stability with a self-organizing neural network. in Proceedings of the International Conference on Intelligent Systems Design and Applications (ISDA 10) 248–253 (IEEE Press, 2010). http://doi.org/10.1109/ISDA.2010.5687256