Better than Deep Learning: Gradient Boosting Machines (GBM) in R - eRum Conference - Budapest, May 2018

Transcript

1. Better than Deep Learning: Gradient Boosting Machines (GBM) in R Szilárd Pafka, PhD Chief Scientist, Epoch (USA) eRum Conference, Budapest May 2018
3. Disclaimer: I am not representing my employer (Epoch) in this talk I cannot confirm nor deny if Epoch is using any of the methods, tools, results etc. mentioned in this talk
4. Source: Andrew Ng
5. Source: Andrew Ng
6. Source: Andrew Ng
11. Source: https://twitter.com/iamdevloper/
13. ...
16. http://www.cs.cornell.edu/~alexn/papers/empirical.icml06.pdf http://lowrank.net/nikos/pubs/empirical.pdf
17. http://www.cs.cornell.edu/~alexn/papers/empirical.icml06.pdf http://lowrank.net/nikos/pubs/empirical.pdf
18. structured/tabular data: GBM (or RF) very small data: LR very large sparse data: LR with SGD (+L1/L2) images/videos, speech: DL
19. structured/tabular data: GBM (or RF) very small data: LR very large sparse data: LR with SGD (+L1/L2) images/videos, speech: DL it depends
20. structured/tabular data: GBM (or RF) very small data: LR very large sparse data: LR with SGD (+L1/L2) images/videos, speech: DL it depends / try them all
21. structured/tabular data: GBM (or RF) very small data: LR very large sparse data: LR with SGD (+L1/L2) images/videos, speech: DL it depends / try them all / hyperparam tuning
22. structured/tabular data: GBM (or RF) very small data: LR very large sparse data: LR with SGD (+L1/L2) images/videos, speech: DL it depends / try them all / hyperparam tuning / ensembles
23. structured/tabular data: GBM (or RF) very small data: LR very large sparse data: LR with SGD (+L1/L2) images/videos, speech: DL it depends / try them all / hyperparam tuning / ensembles feature engineering
24. structured/tabular data: GBM (or RF) very small data: LR very large sparse data: LR with SGD (+L1/L2) images/videos, speech: DL it depends / try them all / hyperparam tuning / ensembles feature engineering / other goals e.g. interpretability
25. structured/tabular data: GBM (or RF) very small data: LR very large sparse data: LR with SGD (+L1/L2) images/videos, speech: DL it depends / try them all / hyperparam tuning / ensembles feature engineering / other goals e.g. interpretability the title of this talk was misguided
26. structured/tabular data: GBM (or RF) very small data: LR very large sparse data: LR with SGD (+L1/L2) images/videos, speech: DL it depends / try them all / hyperparam tuning / ensembles feature engineering / other goals e.g. interpretability the title of this talk was misguided but so is recently almost every use of the term AI
27. Source: Hastie etal, ESL 2ed
28. Source: Hastie etal, ESL 2ed
29. Source: Hastie etal, ESL 2ed
30. Source: Hastie etal, ESL 2ed
33. 10x
36. 10x
48. http://www.jmlr.org/papers/volume13/bergstra12a/bergstra12a.pdf
49. http://www.argmin.net/2016/06/20/hypertuning/
55. More:
58. Backup Slides
59. All benchmarks are wrong, but some are useful