Return on Investment in Machine Learning: Crossing the Chasm between Academia and Business

• Autorzy: Mizgajski Jan , Szymczak Adrian , Morzy Mikołaj , Augustyniak Łukasz , Szymański Piotr , Żelasko Piotr

Identyfikatory

DOI

10.2478/fcds-2020-0015

• Języki publikacji: EN

Abstrakty

EN

Academia remains the central place of machine learning education. While academic culture is the predominant factor influencing the way we teach machine learning to students, many practitioners question this culture, claiming the lack of alignment between academic and business environments. Drawing on professional experiences from both sides of the chasm, we describe the main points of contention, in the hope that it will help better align academic syllabi with the expectations towards future machine learning practitioners. We also provide recommendations for teaching of the applied aspects of machine learning.

Słowa kluczowe

EN

applied machine learning; machine learning teaching; machine learning engineering

• Wydawca: Wydawnictwo Politechniki Poznańskiej
• Czasopismo: Foundations of Computing and Decision Sciences
• Rocznik: 2020
• Tom: Vol. 45, No. 4
• Strony: 281--304
• Opis fizyczny: Bibliogr. 119 poz.

Twórcy

autor

Mizgajski Jan

• Avaya

autor

Szymczak Adrian

• Avaya

autor

Morzy Mikołaj

• Poznan University of Technology

autor

Augustyniak Łukasz

• Avaya
• Wroclaw University of Science and Technology

autor

Szymański Piotr

• Avaya
• Wroclaw University of Science and Technology

autor

Żelasko Piotr

• The Johns Hopkins University

Bibliografia

• [1] Algorithmia. algorithmia.com/product. Accessed: 2020-11-30.
• [2] Amazon ground truth. https://amzn.to/3g0AGqf. Accessed: 2020-11-30.
• [3] Amazon mechanical turk. www.mturk.com. Accessed: 2020-11-30.
• [4] Amazon open data. registry.opendata.aws. Accessed: 2020-11-30.
• [5] Breaking the 80/20 rule: How data catalogs transform data scientists’ productivity. ibm.co/2QhuGxK. Accessed: 2020-11-30.
• [6] Cleaning big data: Most time-consuming, least enjoyable data science task, survey says. bit.ly/3d3M0zQ. Accessed: 2020-11-30.
• [7] Cookiecutter data science. https://bit.ly/3msmp8j. Accessed: 2020-11-30.
• [8] Data gov. www.data.gov. Accessed: 2020-11-30.
• [9] Data usa. datausa.io. Accessed: 2020-11-30.
• [10] Data version control. dvc.org. Accessed: 2020-11-30.
• [11] Figure eight. www.figure-eight.com. Accessed: 2020-11-30.
• [12] Flair. github.com/flairNLP/flair. Accessed: 2020-11-30.
• [13] Ganlab. poloclub.github.io/ganlab. Accessed: 2020-11-30.
• [14] Google dataset search. https://bit.ly/2JucsbU. Accessed: 2020-11-30.
• [15] Huggingface. https://bit.ly/39vUDE4. Accessed: 2020-11-30.
• [16] Jupyanno. github.com/chestrays/jupyanno. Accessed: 2020-11-30.
• [17] Kaggle datasets. www.kaggle.com/datasets. Accessed: 2020-11-30.
• [18] Kedro. github.com/quantumblacklabs/kedro. Accessed: 2020-11-30.
• [19] Metaflow. metaflow.org. Accessed: 2020-11-30.
• [20] Mlflow. mlflow.org. Accessed: 2020-11-30.
• [21] Neptune. neptune.ml. Accessed: 2020-11-30.
• [22] Network repository. networkrepository.com. Accessed: 2020-11-30.
• [23] Notion: All-in-one workplace. ww.notion.so. Accessed: 2020-11-30.
• [24] Pigeon. github.com/agermanidis/pigeon. Accessed: 2020-11-30.
• [25] Sagemaker. aws.amazon.com/sagemaker. Accessed: 2020-11-30.
• [26] The staggering cost of training sota ai models. bit.ly/39O20nL. Accessed: 2020-11-30.
• [27] Tensorboard. www.tensorflow.org/tensorboard. Accessed: 2020-11-30.
• [28] Tf serving. github.com/tensorflow/serving. Accessed: 2020-11-30.
• [29] Tldrlegal - software licenses explained in plain english. tldrlegal.com/. Accessed: 2020-11-30.
• [30] Uci ml repository. archive.ics.uci.edu/ml. Accessed: 2020-11-30.
• [[31] Visdom. github.com/facebookresearch/visdom. Accessed: 2020-11-30.
• [[32] Weights & biases. www.wandb.com. Accessed: 2020-11-30.
• [33] Yellowbrick. www.scikit-yb.org. Accessed: 2020-11-30.
• [34] Fair crowd work: Shedding light on the real work of crowd-, platform-, and app- based work. http://faircrowd.work/platform-reviews, 2018. Accessed: 2020-11-30.
• [35] Amershi S., Begel A., Bird C., DeLine R., Gall H., Kamar E., Nagappan N., Nushi B., and Zimmermann T. Software engineering for machine learning: A case study. In Ąlst IEEE/ACM International Conference on Software Engineering: Software Engineering in Practice, pages 291-300. IEEE, 2019.
• [36] Badene S., Thompson K., Lorre J.-P., and Asher N. Weak supervision for learning discourse structure. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 2296-2305, 2019.
• [37] Bernardi L., Mavridis T., and Estevez P. 150 successful machine learning models: 6 lessons learned at booking.com. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pages 1743-1751, 2019.
• [38] Bloice M.D., Roth P.M., and Holzinger A. Biomedical image augmentation using augmentor. Bioinformatics, 35(21):4522-4524, 2019.
• [39] Bolukbasi T., Chang K.-W., Zou J., Saligrama V., and Kalai A. Man is to computer programmer as woman is to homemaker? debiasing word embeddings, 2016.
• [40] Bosch J., Crnkovic I., and Olsson H.H. Engineering ai systems: A research agenda, 2020.
• [41] Breck E., Cai S., Nielsen E., Salib M., and Sculley D. What’s your ml test score? a rubric for ml production systems. In Reliable Machine Learning in the Wild - NIPS 2016 Workshop, 2016.
• [42] Breiman L. Random forests. Machine Learning, 45(1):5-32, 2001.
• [43] Brown T.B., Mann B., Ryder N., Subbiah M., Kaplan J., Dhariwal P., Neelakantan A., Shyam P., Sastry G., Askell A., Agarwal S., Herbert-Voss A., Krueger G., Henighan T., Child R., Ramesh A., Ziegler D.M., Wu J., Winter C., Hesse C., Chen M., Sigler E., Litwin M., Gray S., Chess B., Clark J., Berner C., McCandlish S., Radford A., Sutskever I., and Amodei D. Language models are few-shot learners, 2020.
• [44] Buslaev A., Iglovikov V.I., Khvedchenya E., Parinov A., Druzhinin M., and Kalinin A.A. Albumentations: fast and flexible image augmentations. Information, 11(2):125, 2020.
• [45] Byun T. and Rayadurgam S. Manifold for machine learning assurance. Proceedings of the ACM/IEEE 42nd International Conference on Software Engineering, 06 2020.
• [46] Dehghani M., Severyn A., Rothe S., and Kamps J. Learning to learn from weak supervision by full supervision, 2017.
• [47] Demir S., Eniser H.F., and Sen A. Deepsmartfuzzer: Reward guided test generation for deep learning, 2019.
• [48] Devlin J., Chang M.-W., Lee K., and Toutanova K. Bert: Pretraining of deep bidirectional transformers for language understanding, 2019.
• [49] Dignum V. Responsible artificial intelligence: designing ai for human values. The ITU Journal on Future and Evolving Technologies, 2017.
• [50] Dingwall N. and Potts C. Mittens: An extension of glove for learning domain-specialized representations, 2018.
• [51] Doersch C. and Zisserman A. Multi-task self-supervised visual learning. In Proceedings of the IEEE International Conference on Computer Vision, pages 2051-2060, 2017.
• [52] Drozdowski M., Kowalski D., Mizgajski J., Mokwa D., and Pawlak G. Mind the gap: a heuristic study of subway tours. Journal of Heuristics, 20(5):561-587, 2014.
• [53] Fayyad U., Piatetsky-Shapiro G., and Smyth P. From data mining to knowledge discovery in databases. AI magazine, 17(3):37-37, 1996.
• [54] Gagolewski M. et al. Benchmark suite for clustering algorithms - version 1, 2020.
• [55] Garcia M. Racist in the machine: The disturbing implications of algorithmic bias. World Policy Journal, 33(4):111-117, 2016.
• [56] Gerasimou S., Eniser H.F., Sen A., and Cakan A. Importance-driven deep learning system testing, 2020.
• [57] Gilyazev R. and Turdakov D.Y. Active learning and crowdsourcing: A survey of optimization methods for data labeling. Programming and Computer Software, 44(6):476-491, 2018.
• [58] Gofman M. and Jin Z. Artificial intelligence, human capital, and innovation. Human Capital, and Innovation (August 20, 2019), 2019.
• [59] Gong C., Zhang H., Yang J., and Tao D. Learning with inadequate and incorrect supervision. In 2017 IEEE International Conference on Data Mining (ICDM), pages 889-894. IEEE, 2017.
• [60] Gottgtroy P. Ontology driven knowledge discovery process: a proposal to integrate ontology engineering and kdd. PACIS 2007 Proceedings, page 88, 2007.
• [61] Hadiwinoto C. and Ng H.T. Upping the ante: Towards a better benchmark for chinese-to-english machine translation, 2018.
• [62] Hager G.D., Drobnis A., Fang F., Ghani R., Greenwald A., Lyons T., Parkes D.C., Schultz J., Saria S., Smith S.F., and Tambe M. Artificial intelligence for social good, 2019.
• [63] Hand D.J. and Khan S. Validating and verifying ai systems. Patterns, 1(3):100037, 2020.
• [64] Hao D., Zhang L., Sumkin J., Mohamed A., and Wu S. Inaccurate labels in weakly supervised deep learning: Automatic identification and correction and their impact on classification performance. IEEE Journal of Biomedical and Health Informatics, 2020.
• [65] Hendrycks D., Mazeika M., Kadavath S., and Song D. Using self-supervised learning can improve model robustness and uncertainty. In Advances in Neural Information Processing Systems, pages 15663-15674, 2019.
• [66] Ho D., Liang E., Chen X., Stoica I., and Abbeel P. Population based augmentation: Efficient learning of augmentation policy schedules. In Proceedings of the 36th International Conference on Machine Learning, volume 97 of Proceedings of Machine Learning Research, pages 2731-2741, Long Beach, California, USA, 06 2019. PMLR.
• [67] Honnibal M. and Montani I. spaCy 2: Natural language understanding with Bloom embeddings, convolutional neural networks and incremental parsing. 2017.
• [68] Howard A. and Borenstein J. The ugly truth about ourselves and our robot creations: The problem of bias and social inequity. Science and engineering ethics, 24(5):1521-1536, 2018.
• [69] Howard A.G., Zhu M., Chen B., Kalenichenko D., Wang W., Weyand T., Andreetto M., and Adam H. Mobilenets: Efficient convolutional neural networks for mobile vision applications, 2017.
• [70] Howard J. and Ruder S. Universal language model fine-tuning for text classification, 2018.
• [71] Inmon B. Data Lake Architecture: Designing the Data Lake and avoiding the garbage dump. Technics Publications, 2016.
• [72] Inmon W.H. Building the data warehouse. John Wiley & Sons, 2005.
• [73] Kessler J.S. Scattertext: a browser-based tool for visualizing how corpora differ, 2017.
• [74] Kim M., Zimmermann T., DeLine R., and Begel A. Data scientists in software teams: State of the art and challenges. IEEE Transactions on Software Engineering, 44(11):1024-1038, 2017.
• [75] Krizhevsky A., Sutskever I., and Hinton G.E. Imagenet classification with deep convolutional neural networks. Communications of the ACM, 60(6):84-90, 2017.
• [76] Lacoste A., Luccioni A., Schmidt V., and Dandres T. Quantifying the carbon emissions of machine learning, 2019.
• [77] Lazer D., Kennedy R., King G., and Vespignani A. The parable of google flu: traps in big data analysis. Science, 343(6176):1203-1205, 2014.
• [78] Ledwich M. and Zaitsev A. Algorithmic extremism: Examining youtube’s rabbit hole of radicalization, 2019.
• [79] Lee S.W. and Kerschberg L.A methodology and life cycle model for data mining and knowledge discovery in precision agriculture. In SMC’98 Conference Proceedings. 1998 IEEE International Conference on Systems, Man, and Cybernetics, volume 3, pages 2882-2887, 1998.
• [80] Ma E. NLP augmentation. github.com/makcedward/nlpaug, 2019.
• [81] Matignon R. Data mining using SAS enterprise miner, volume 638. John Wiley & Sons, 2007.
• [82] Mikolov T., Chen K., Corrado G., and Dean J. Efficient estimation of word representations in vector space, 2013.
• [83] Mizgajski J., Szymczak A., Żelasko P., Morzy M., Augustyniak Ł., and Szymański P. Return of investment in machine learning: Crossing the chasm between academia and business. In Proceedings of the PP-RAI’2019 Conference: Polskie Porozumienie na Rzecz Rozwoju Sztucznej Inteligencji, pages 285-291. Wroclaw University of Science and Technology, 2019.
• [84] Montani I. and Honnibal M. Prodigy: A new annotation tool for radically efficient machine teaching. prodi.gy/, 2018. Accessed: 2020-11-30.
• [85] Nakayama H., Kubo T., Kamura J., Taniguchi Y., and Liang X. Doccano: Text annotation tool for human. github.com/doccano/doccano, 2018. Accessed: 2020-11-30.
• [86] Obermeyer Z. and Mullainathan S. Dissecting racial bias in an algorithm that guides health decisions for 70 million people. In Proceedings of the Conference on Fairness, Accountability, and Transparency, pages 89-89, 2019.
• [87] Paszke A., Gross S., Massa F., Lerer A., Bradbury J., Chanan G., Killeen T., Lin Z., Gimelshein N., Antiga L., et al. Pytorch: An imperative style, high-performance deep learning library. Advances in Neural Information Processing Systems, 32:8026-8037, 2019.
• [88] Pennington J., Socher R., and Manning C.D. Glove: Global vectors for word representation. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1532-1543, 2014.
• [89] Purdy M. and Daugherty P. Why artificial intelligence is the future of growth, 2016.
• [90] Pustejovsky J. and Stubbs A. Natural Language Annotation for Machine Learning: A Guide to Corpus-Building for Applications. O’Reilly Media, 2012.
• [91] Ratner A., Bach S.H., Ehrenberg H., Fries J., Wu S., and Re C. Snorkel: rapid training data creation with weak supervision. Proceedings of the VLDB Endowment, 11(3):269-282, 11 2017.
• [92] Ribeiro M.T., Wu T., Guestrin C., and Singh S. Beyond accuracy: Behavioral testing of nlp models with checklist, 2020.
• [93] Roh Y., Heo G., and Whang S.E. A survey on data collection for machine learning: a big data - ai integration perspective, 2019.
• [94] Ruder S., Peters M.E., Swayamdipta S., and Wolf T. Transfer learning in natural language processing. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Tutorials, pages 15-18, 2019.
• [95] Schwartz R., Dodge J., Smith N.A., and Etzioni O. Green ai, 2019.
• [96] Sculley D., Holt G., Golovin D., Davydov E., Phillips T., Ebner D., Chaud-hary V., Young M., Crespo J.-F., and Dennison D. Hidden technical debt in machine learning systems. In Advances in Neural Information Processing Systems, pages 2503-2511, 2015.
• [97] Settles B. Active learning literature survey. Technical report, University of Wisconsin-Madison Department of Computer Sciences, 2009.
• [98] Shane J. You Look Like a Thing and I Love You: How Artificial Intelligence Works and why It’s Making the World a Weirder Place. Voracious, 2019.
• [99] Shi Z.R., Wang C., and Fang F. Artificial intelligence for social good: A survey, 2020.
• [100] Simonyan K. and Zisserman A. Very deep convolutional networks for large-scale image recognition, 2015.
• [101] Strobelt H., Gehrmann S., Behrisch M., Perer A., Pfister H., and Rush A.M. Seq2seq-vis: A visual debugging tool for sequence-to-sequence models. IEEE Transactions on Visualization and Computer Graphics, 25(1):353-363, 2018.
• [102] Sun Z., Zhang J.M., Harman M., Papadakis M., and Zhang L. Automatic testing and improvement of machine translation, 2019.
• [103] Tatman R. Gender and dialect bias in youtube’s automatic captions. In Proceedings of the First ACL Workshop on Ethics in Natural Language Processing, pages 53-59, 2017.
• [104] Tufekci Z. Youtube’s recommendation algorithm has a dark side. bit.ly/ 2m09tvZ. Accessed: 2020-11-30.
• [105] Van Engelen J.E. and Hoos H.H. A survey on semi-supervised learning. Machine Learning, 109(2):373-440, 2020.
• [106] VanderPlas J. The big data brain drain: Why science is in trouble. Accessed: 2020-11-30.
• [107] VanderPlas J. Hacking academia: Data science and the university. Accessed: 2020-11-30.
• [108] Vassiliadis P. A survey of extract-transform-load technology. International Journal of Data Warehousing and Mining, 5(3):1-27, 2009.
• [109] Wagstaff K. Machine learning that matters, 2012.
• [110] Wan Z., Xia X., Lo D., and Murphy G.C. How does machine learning change software development practices? IEEE Transactions on Software Engineering, pages 1-14, 2019.
• [111] Wang J., Gou L., Shen H.-W., and Yang H. Dqnviz: A visual analytics approach to understand deep q-networks. IEEE Transactions on Visualization and Computer Graphics, 25(1):288-298, 2018.
• [112] Weiss K., Khoshgoftaar T.M., and Wang D. A survey of transfer learning. Journal of Big data, 3(1):9, 2016.
• [113] Wirth R. and Hipp J. Crisp-dm: Towards a standard process model for data mining. In Proceedings of the Ąth International Conference on the Practical Applications of Knowledge Discovery and Data Mining, pages 29-39. Springer-Verlag, 2000.
• [114] Yan M., Wang L., and Fei A. Artdl: Adaptive random testing for deep learning systems. IEEE Access, 2019.
• [115] Yosinski J., Clune J., Nguyen A., Fuchs T., and Lipson H. Understanding neural networks through deep visualization, 2015.
• [116] Zafrir O., Boudoukh G., Izsak P., and Wasserblat M. Q8bert: Quantized 8bit bert, 2019.
• [117] Zeiler M.D. and Fergus R. Visualizing and understanding convolutional networks. In 13th European Conference on Computer Vision, ECCV 2014, pages 818-833. Springer Verlag, 2014.
• [118] Zhang J.M., Harman M., Ma L., and Liu Y. Machine learning testing: Survey, landscapes and horizons. IEEE Transactions on Software Engineering, 2020.
• [119] Zinkevich M. Rules of machine learning: Best practices for ml engineering. developers.google.com/machine-learning/guides/rules-of-ml.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2020).