From a Developer's POV: is Machine Learning Reshaping the World? - Simone Scardapane - Codemotion Rome 2017
0 likes519 views
The document discusses the impact of machine learning on various fields, including software trends like simpler ML libraries, feasible deep learning, and machine learning as a service. It highlights challenges, such as the technical debt associated with real-world ML systems and the implications for programmers. The content draws upon various studies and reports to illustrate current developments and considerations in the machine learning landscape.
![ROME 24-25 MARCH 2017
PathNet: Evolution Channels Gradient Descent in Super Neural
Networks [arXiv preprint]](https://image.slidesharecdn.com/ks54pdfrrqifnv1sgzfs-signature-5e60f00ef9cb0f2acb38cf17199ee04ae4edcf8fdd3dfe78961880780aed8435-poli-170330155328/85/From-a-Developer-s-POV-is-Machine-Learning-Reshaping-the-World-Simone-Scardapane-Codemotion-Rome-2017-24-320.jpg)
![ROME 24-25 MARCH 2017
DeepCoder: Learning to Write Programs [arXiv preprint]](https://image.slidesharecdn.com/ks54pdfrrqifnv1sgzfs-signature-5e60f00ef9cb0f2acb38cf17199ee04ae4edcf8fdd3dfe78961880780aed8435-poli-170330155328/85/From-a-Developer-s-POV-is-Machine-Learning-Reshaping-the-World-Simone-Scardapane-Codemotion-Rome-2017-38-320.jpg)
From a Developer's POV: is Machine Learning Reshaping the World? - Simone Scardapane - Codemotion Rome 2017
- 1. From a Developer's POV: is Machine Learning Reshaping the World? Simone Scardapane ROME 24-25 MARCH 2017
- 2. ROME 24-25 MARCH 2017 The age of analytics: Competing in a data- driven world (McKinsey Report)
- 3. ROME 24-25 MARCH 2017
- 4. ROME 24-25 MARCH 2017
- 5. ROME 24-25 MARCH 2017When do you need machine learning? Two big challenges in machine learning (Léon Bottou)
- 6. ROME 24-25 MARCH 2017Software trend 1: simpler ML libraries
- 7. ROME 24-25 MARCH 2017Can we predict the skill level of a player?
- 8. ROME 24-25 MARCH 2017 Thompson, J.J., Blair, M.R., Chen, L. and Henrey, A.J., 2013. Video game telemetry as a critical tool in the study of complex skill learning. PloS one, 8(9), p.e75129.
- 9. ROME 24-25 MARCH 2017
- 10. ROME 24-25 MARCH 2017
- 11. ROME 24-25 MARCH 2017
- 12. ROME 24-25 MARCH 2017We still need to choose a classifier! :-(
- 13. ROME 24-25 MARCH 2017Auto Machine Learning An overview of the AutoML system taken from: Feurer, M., Klein, A., Eggensperger, K., Springenberg, J., Blum, M. and Hutter, F., 2015.Efficient and robust automated machine learning. In Advances in Neural Information Processing Systems (pp. 2962- 2970).
- 14. ROME 24-25 MARCH 2017Software trend 2: feasible deep learning
- 15. ROME 24-25 MARCH 2017Can we distinguish between them? Labeled faces in the wild
- 16. ROME 24-25 MARCH 2017
- 17. ROME 24-25 MARCH 2017
- 18. ROME 24-25 MARCH 2017
- 19. ROME 24-25 MARCH 2017
- 20. ROME 24-25 MARCH 2017
- 21. ROME 24-25 MARCH 2017
- 22. ROME 24-25 MARCH 2017
- 23. ROME 24-25 MARCH 2017Software trend 3: MODULAR deep learning
- 24. ROME 24-25 MARCH 2017 PathNet: Evolution Channels Gradient Descent in Super Neural Networks [arXiv preprint]
- 25. ROME 24-25 MARCH 2017Software trend 4: machine learning as a service
- 26. ROME 24-25 MARCH 2017
- 27. ROME 24-25 MARCH 2017
- 28. ROME 24-25 MARCH 2017Software trend 5: reinforcement learning repositories
- 29. ROME 24-25 MARCH 2017 Universe (OpenAI) DeepMind Lab (Google DeepMind)
- 30. ROME 24-25 MARCH 2017Some words of caution
- 31. ROME 24-25 MARCH 2017 Breaking linear classifiers on ConvNet
- 32. ROME 24-25 MARCH 2017 McDaniel, P., Papernot, N. and Celik, Z.B., 2016. Machine learning in adversarial settings. IEEE Security & Privacy, 14(3), pp. 68-72.
- 33. ROME 24-25 MARCH 2017The technical debt of machine learning
- 34. ROME 24-25 MARCH 2017 Machine learning offers a fantastically powerful toolkit for building useful complex prediction systems quickly. ... it is dangerous to think of these quick wins as coming for free. ... it is common to incur massive ongoing maintenance costs in real- world ML systems. Hidden Technical Debt in Machine Learning Systems (NIPS 2015)
- 35. ROME 24-25 MARCH 2017
- 36. ROME 24-25 MARCH 2017
- 37. ROME 24-25 MARCH 2017Will machine learning replace programmers?
- 38. ROME 24-25 MARCH 2017 DeepCoder: Learning to Write Programs [arXiv preprint]
- 39. ROME 24-25 MARCH 2017
- 40. ROME 24-25 MARCH 2017
- 41. ROME 24-25 MARCH 2017
- 42. ROME 24-25 MARCH 2017
- 43. ROME 24-25 MARCH 2017