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
27

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
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
16
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
25
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
6
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
7
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
79
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
18
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
50
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
17
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
6

Featured

See All Featured
5ce91fa49a613cbc3e20d5f96856473f?s=48 edds
56
9.3k
E43f8dfd01057f8304f5f8fb9a294d7d?s=48 jeffersonlam
328
15k
Ba530e786553390f3fb271848485681c?s=48 3n
163
22k
245cee81a9c424266e5e401d844ea881?s=48 lara
172
9.5k
5b9fe87ec1faa67a4599782930f45ec9?s=48 sstephenson
144
12k
766b9746b59e6f09e280cd33cf4ed419?s=48 skipperchong
7
660
E190023b66e2b8aa73a842b106920c93?s=48 zenorocha
297
39k
61857dafbd287b3027c4dcea9008ad3c?s=48 carmenhchung
26
1.3k
5f83ef806155b403c3393160ce51f955?s=48 jlugia
216
16k
Af6a12710770ad9a7cfd08c97efd40d3?s=48 pedronauck
652
110k
1b75d89772b7eea1f6c89fbf362607d9?s=48 moore
125
21k
E6c6e133e74c3b83f04d2861deaa1c20?s=48 searls
204
35k

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