Slide 1
Slide 1 text
Machine Learning for
Gophers
@francesc
Slide 2
Slide 2 text
1996 - Exciting times
Slide 3
Slide 3 text
1996 - Exciting times (at least for computers)
Slide 4
Slide 4 text
2016 - Very exciting times
Slide 5
Slide 5 text
2016 - Very exciting times
Slide 6
Slide 6 text
name elo rating matches
Fan Hui 3,144 5:0 against Fan Hui
Lee Sedol 3,739 4:1 against Lee Sedol
Master 4,858 60:0 against professional players
Zero 5,185 100:0 against AlphaGo Lee
89:11 against AlphaGo Master
training
hardware
176 GPU
48 TPU
4 TPU
4 TPU
2017 - E_TOO_EXCITE
Slide 7
Slide 7 text
deepmind.com/blog/alphago-zero-learning-scratch
Slide 8
Slide 8 text
What is Machine Learning?
“A computer program is said to learn from experience E, with respect to
some task T, and some performance measure P, if its performance on T as
measured by P improves with experience E”
- Tom Mitchell
Slide 9
Slide 9 text
What is Machine Learning?
“A computer program is said to learn from experience E, with respect to
some task T, and some performance measure P, if its performance on T as
measured by P improves with experience E”
- Tom Mitchell
Slide 10
Slide 10 text
What is Machine Learning?
“A computer program is said to learn from experience E, with respect to
some task T, and some performance measure P, if its performance on T as
measured by P improves with experience E”
- Tom Mitchell
Slide 11
Slide 11 text
What is Machine Learning?
Slide 12
Slide 12 text
What is Machine Learning?
● task: recognizing handwritten digits
● performance measure: percentage of digits labeled correctly
● experience: try to predict, fail, improve it, repeat
Slide 13
Slide 13 text
Input
0 0 0 0 0 0 0 0
0 0 255 0 0 255 0 0
0 0 255 0 0 255 0 0
0 0 255 255 255 255 0 0
0 0 0 0 0 255 0 0
0 0 0 0 0 255 0 0
0 0 0 0 0 255 0 0
0 0 0 0 0 0 0 0
28 x 28
, 4
Slide 14
Slide 14 text
● Very simple technique
y = X * W + b
Logistic regression
input biases
weights
prediction
Slide 15
Slide 15 text
● Average of how “off” predictions were
J = -1/m * sum(dist(pred, label))
Cost function
Slide 16
Slide 16 text
● Finding the W and b that minimize J
● We know what direction to go by deriving J vs W and b
not gonna show that
Optimizing
Slide 17
Slide 17 text
mnist tutorial?
Slide 18
Slide 18 text
Can we do this in Go?
Slide 19
Slide 19 text
● Reading images
● Multiplying matrices
● Visualization of datasets
● Making it fast
What do we need?
Slide 20
Slide 20 text
● matrices and linear algebra
● statistics
● probability distributions
● differentiation, integration, and optimization
● plotting
● and much more!
gonum
Slide 21
Slide 21 text
1
2m
(X * theta - y)2
Σ
Slide 22
Slide 22 text
gonum
func ComputeCost(X, y, theta *mat.Dense) float64 {
m, _ := X.Dims()
h := new(mat.Dense)
h.Mul(X, theta)
h.Sub(h, y)
h.MulElem(h, h)
return mat.Sum(h) / float64(2*m)
}
Slide 23
Slide 23 text
No content
Slide 24
Slide 24 text
Let’s reinvent the wheel!
github.com/campoy/mat
● focus on API over performance
● immutable matrices
● only float64 matrices
Slide 25
Slide 25 text
ComputeCost with github.com/campoy/mat
func ComputeCost(X, y, theta mat.Matrix) float64 {
h := mat.Minus(mat.Product(X, theta), y)
return mat.Product(h, h).Sum() / float64(2 * X.Rows())
}
Slide 26
Slide 26 text
demo time! *
* not guaranteed to work
Slide 27
Slide 27 text
References
github.com/campoy/mat
github.com/campoy/goml
github.com/campoy/justforfunc
github.com/campoy/tools/imgcat
Slide 28
Slide 28 text
Merci!
@francesc
VP of Developer Relations at source{d}
sourced.tech