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
24

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

Want more?

Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
48
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
15
Ce8e94cc306ba164175f693fb01aa8b0?s=48 szilard
0
36
3ab1249be442027903e1180025340b3f?s=48 reverentgeek
7
340
A9a491b0fcbe0fbce3d64063a37add99?s=48 rmw
1
200
766b9746b59e6f09e280cd33cf4ed419?s=48 skipperchong
0
130
61857dafbd287b3027c4dcea9008ad3c?s=48 carmenhchung
6
260
06f8b41980eb4c577fa40c41d5030c19?s=48 keathley
5
190
C0390e8b11079f8761c0cd3274ed61cb?s=48 productmarketing
3
280
C35e01544a0dd94a2cf1619ee8a42ebb?s=48 clairelew
3
520
6bb82b13cda86d3b821fa44100335ddf?s=48 thoeni
1
140
B3ace07412a5178ea778e7eec161fc0a?s=48 jcasabona
4
230

Transcript

  1. 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. 2.
    None
  3. 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. 4.

    Source: Andrew Ng

  5. 5.

    Source: Andrew Ng

  6. 6.

    Source: Andrew Ng

  7. 7.
    None
  8. 8.
    None
  9. 9.
    None
  10. 10.
    None
  11. 11.

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

  12. 12.
    None
  13. 13.
    None
  14. 14.

    ...

  15. 15.
    None
  16. 16.
    None
  17. 17.
    None
  18. 18.

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

  19. 19.

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

  20. 20.
    None
  21. 21.
    None
  22. 22.
    None
  23. 23.
    None
  24. 24.
    None
  25. 25.
    None
  26. 26.
    None
  27. 27.
    None
  28. 28.
    None
  29. 29.
    None
  30. 30.
    None
  31. 31.
    None
  32. 32.
    None
  33. 33.
    None
  34. 34.
    None
  35. 35.
    None
  36. 36.

    Source: Hastie etal, ESL 2ed

  37. 37.

    Source: Hastie etal, ESL 2ed

  38. 38.

    Source: Hastie etal, ESL 2ed

  39. 39.

    Source: Hastie etal, ESL 2ed

  40. 40.
    None
  41. 41.
    None
  42. 42.
    None
  43. 43.
    None
  44. 44.
    None
  45. 45.
    None
  46. 46.

    10x

  47. 47.
    None
  48. 48.
    None
  49. 49.
    None
  50. 50.

    10x

  51. 51.

    10x

  52. 52.
    None
  53. 53.
    None
  54. 54.
    None
  55. 55.
    None
  56. 56.
    None
  57. 57.
    None
  58. 58.
    None
  59. 59.
    None
  60. 60.
    None
  61. 61.
    None
  62. 62.
    None
  63. 63.
    None
  64. 64.
    None
  65. 65.
    None
  66. 66.
    None
  67. 67.
    None
  68. 68.
    None
  69. 69.
    None
  70. 70.
    None
  71. 71.
    None
  72. 72.
    None
  73. 73.
    None
  74. 74.
    None
  75. 75.
    None
  76. 76.
    None
  77. 77.
    None
  78. 78.
    None
  79. 79.
    None
  80. 80.
    None
  81. 81.
    None
  82. 82.
    None
  83. 83.
    None
  84. 84.
    None
  85. 85.
    None
  86. 86.
    None
  87. 87.
    None
  88. 88.
    None
  89. 89.
    None
  90. 90.
    None
  91. 91.
    None
  92. 92.
    None
  93. 93.

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

  94. 94.

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

  95. 95.
    None
  96. 96.
    None
  97. 97.
    None
  98. 98.
    None
  99. 99.
    None
  100. 100.
    None
  101. 101.
    None
  102. 102.

    CPU 1

  103. 103.

    CPU 1 CPU 2

  104. 104.

    CPU 1 CPU 2

  105. 105.

    CPU 1 CPU 2

  106. 106.

    CPU 1 CPU 2

  107. 107.
    None
  108. 108.
    None
  109. 109.
    None
  110. 110.
    None
  111. 111.
    None
  112. 112.
    None
  113. 113.
    None
  114. 114.

    *

  115. 115.
    None
  116. 116.

    no-one is using this crap

  117. 117.

    (2018)

  118. 118.

    (2018)

  119. 119.
    None
  120. 120.

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

  121. 121.

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

  122. 122.

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

  123. 123.

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

  124. 124.

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

  125. 125.

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

  126. 126.
    None
  127. 127.

    More:

  128. 128.
    None
  129. 129.
    None
  130. 130.
    None
  131. 131.
    None
  132. 132.

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

  133. 133.
    None