Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Better than Deep Learning: Gradient Boosting Machines (GBM) / 2019 edition - Budapest R and Data Science Meetups - Budapest, June 2019

Ce8e94cc306ba164175f693fb01aa8b0?s=47 szilard
June 26, 2019
0
48

Better than Deep Learning: Gradient Boosting Machines (GBM) / 2019 edition - Budapest R and Data Science Meetups - Budapest, June 2019

Ce8e94cc306ba164175f693fb01aa8b0?s=128

szilard

June 26, 2019
Tweet

More Decks by szilard

See All by szilard
Gradient Boosting Machines (GBM): From Zero to Hero (with R and Python Code) - Data Con LA - Oct 2020
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
46
Make Machine Learning Boring Again: Best Practices for Using Machine Learning in Businesses - Albuquerque Machine Learning Meetup (Online) - Aug 2020
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
58
Better than Deep Learning: Gradient Boosting Machines (GBM) - eRum conference - invited talk - June 2020
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
41
Gradient Boosting Machines (GBM): From Zero to Hero (with R and Python Code) - LA Data Science Meetup - February 2020
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
31
A Random Walk in Data Science and Machine Learning in Practice - CEU, Business Analytics Masters - Budapest, Febr 2020
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
130
Better than My Meetup/Conference Talks: Going Deeper in Various GBM Topics - GBM Advanced Workshop - Budapest, Nov 2019
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
37
Gradient Boosting Machines (GBM): From Zero to Hero (with R and Python Code) - Budapest BI Forum, Budapest, Nov 2019
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
96
Make Machine Learning Boring Again: Best Practices for Using Machine Learning in Businesses - LA Data Science Meetup - Playa Vista, August 2019
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
39
Better than Deep Learning: Gradient Boosting Machines (GBM) / 2019 edition - LA R Meetup - Santa Monica, May 2019
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
12

Featured

See All Featured
Designing the Hi-DPI Web
C157b16234f1e75e8eac3698c1d4414a?s=48 ddemaree
272
32k
How to train your dragon (web standard)
053e75a5b48b44d6dd0612795dfb326d?s=48 notwaldorf
60
3.9k
4 Signs Your Business is Dying
E4f5d494ebdc9e7cce1aecf3ce3e8bc1?s=48 shpigford
169
20k
What the flash - Photography Introduction
5ce91fa49a613cbc3e20d5f96856473f?s=48 edds
63
10k
Thoughts on Productivity
A16eb159db895e3b01d3dc95767ad595?s=48 jonyablonski
44
2.4k
Designing for humans not robots
D36d2c1821af9249b69ff7f5ed60529b?s=48 tammielis
242
24k
Documentation Writing (for coders)
61857dafbd287b3027c4dcea9008ad3c?s=48 carmenintech
48
2.6k
Automating Front-end Workflow
96270e4c3e5e9806cf7245475c00b275?s=48 addyosmani
1351
200k
Principles of Awesome APIs and How to Build Them.
B47b288784fda77a94aeb234ca743c24?s=48 keavy
113
15k
Art Directing for the Web. Five minutes with CSS Template Areas
433acaea1012b25d97ae66da9b998dad?s=48 malarkey
196
9.5k
Designing with Data
F57e2136eb9895eb95ff5dc8a1c02677?s=48 zakiwarfel
91
4k
Making the Leap to Tech Lead
32de0bd2ba869609d26fd052a4622778?s=48 cromwellryan
113
7.4k

Transcript

  1. Better than Deep Learning: Gradient Boosting Machines (GBM) - 2019

    edition Szilard Pafka, PhD Chief Scientist, Epoch (USA) Budapest R & Data Science Meetups June 2019
  2. None

  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

  7. None
  8. None
  9. None
  10. None

  11. Source: https://twitter.com/iamdevloper/

  12. None
  13. None

  14. ...

  15. None
  16. None
  17. None

  18. http://lowrank.net/nikos/pubs/empirical.pdf http://www.cs.cornell.edu/~alexn/papers/empirical.icml06.pdf

  19. http://lowrank.net/nikos/pubs/empirical.pdf http://www.cs.cornell.edu/~alexn/papers/empirical.icml06.pdf

  20. None
  21. None
  22. None
  23. None
  24. None
  25. None
  26. None
  27. None
  28. None
  29. None
  30. None
  31. None
  32. None
  33. None
  34. None
  35. None

  36. Source: Hastie etal, ESL 2ed

  37. Source: Hastie etal, ESL 2ed

  38. Source: Hastie etal, ESL 2ed

  39. Source: Hastie etal, ESL 2ed

  40. None
  41. None
  42. None
  43. None
  44. None
  45. None

  46. 10x

  47. None
  48. None
  49. None

  50. 10x

  51. 10x

  52. None
  53. None
  54. None
  55. None
  56. None
  57. None
  58. None
  59. None
  60. None
  61. None
  62. None
  63. None
  64. None
  65. None
  66. None
  67. None
  68. None
  69. None
  70. None
  71. None
  72. None
  73. None
  74. None
  75. None
  76. None
  77. None
  78. None
  79. None
  80. None
  81. None
  82. None
  83. None
  84. None
  85. None
  86. None
  87. None
  88. None
  89. None
  90. None
  91. None
  92. None

  93. http://www.jmlr.org/papers/volume13/bergstra12a/bergstra12a.pdf

  94. http://www.argmin.net/2016/06/20/hypertuning/

  95. None
  96. None
  97. None
  98. None
  99. None
  100. None
  101. None

  102. CPU 1

  103. CPU 1 CPU 2

  104. CPU 1 CPU 2

  105. CPU 1 CPU 2

  106. CPU 1 CPU 2

  107. None
  108. None
  109. None
  110. None
  111. None
  112. None
  113. None

  114. *

  115. None

  116. no-one is using this crap

  117. (2018)

  118. (2018)

  119. None

  120. Source: https://www.linkedin.com/pulse/winning-solution-kaggledays-2019-competition-san-francisco-mark-peng/

  121. Source: https://www.linkedin.com/pulse/winning-solution-kaggledays-2019-competition-san-francisco-mark-peng/

  122. Source: https://www.linkedin.com/pulse/winning-solution-kaggledays-2019-competition-san-francisco-mark-peng/

  123. Source: https://www.linkedin.com/pulse/winning-solution-kaggledays-2019-competition-san-francisco-mark-peng/

  124. Source: https://www.linkedin.com/pulse/winning-solution-kaggledays-2019-competition-san-francisco-mark-peng/

  125. Source: https://www.linkedin.com/pulse/winning-solution-kaggledays-2019-competition-san-francisco-mark-peng/

  126. None

  127. More:

  128. None
  129. None
  130. None
  131. None

  132. Info and apply here: bit.ly/gbm-bp19

  133. None