Cross Validated
Sponsored by
Sponsored logo
Profile Activity
Meta user
9,700
reputation
1
29
63

Skander H.

AI / ML / Data Science person.

More practitioner than advanced theoretician.

I spend my days helping business people figure out which ML or statistical models they can use to solve their challenges, and my nights wishing Deep Learning could be used with everything, everywhere, all the time (it can't).

169
answers
86
questions
~267k
people reached
  • Seattle, WA, USA
  • Member for 5 years, 4 months
  • 865 profile views
  • Last seen 6 hours ago

Communities (1)

Top Meta Posts
7
1

Top network posts

We respect a laser-like focus on one topic.

Top tags (177)

Score 257
Posts 170
Posts % 67
Score 187
Posts 116
Score 133
Posts 51
Score 71
Posts 42
Score 68
Posts 40
Score 41
Posts 27
View all tags →

Top posts (255)

All Questions Answers
Votes Newest
37 Why doesn't regularization solve Deep Neural Nets hunger for data? May 11 '18
35 How to handle a “self defeating” prediction model? Oct 22 '18
23 Mathematical/Algorithmic definition for overfitting Jul 2 '19
21 Do neural networks learn a function or a probability density function? May 21 '18
20 Is it possible to automate time series forecasting? Dec 6 '18
18 Can someone explain the importance of mean stationarity in time series? Jun 1 '20
18 On George Box, Galit Shmueli and the scientific method? Jun 14 '19
17 If an auto-regressive time series model is non-linear, does it still require stationarity? Apr 9 '18
16 Why is Kullback-Leilbler divergence a better metric for measuring distance between two probability distributions than squared error? Jun 1 '19
13 gradient descent and local maximum Aug 14 '18
View all questions and answers →

Badges (93)

Gold

1

Rarest

Silver

29

Rarest

Bronze

63

Rarest

View all badges →