ABCDEFGHIJKLMNOPQRSTUVWXYZ
1
PartBlockContentReadingAssignment dueNo of pages to read
2
I1Introduction to supervised learning, logistic and linear regression, objective function, gradient descent, stochastic gradient descent. Evaluation, Bias and Variance trade-off, generalization error, training error, cross-validation, regularization, ridge regression and LASSO, probabilistic interpretation of regularizationEfron (2020): Section 1-3,5-6,8
Ruder (2016): Section 1-4 (skip 4.7 and 4.8)
DL: 5-5.3 (5.1.4 can be skipped)
DL (optional): 4.3 (skip 4.3.1), 5.1.4, 5.9, 8.3, 8.5
ESL: 3.4-3.4.3, 4.4-4.4.1, 4.4.4, 7-7.6, 7.10, 7.12
DLR (optional): 4 - 4.4 (skip 4.4.1-)
Video:
ISLV: Ch. 1, Video 2, Ch. 4: Video 2,3 (repetition of logistic regression), Ch. 5, Video 1-3		1Mandatory: 78p
Optional: 34p
3
2Ensemble methods, bagging, boosting, decision treesESL: 8.7, 9.2, 10 - 10.3, 10.9-10.13,15-15.3.2, 15.3.4, 16-16.2.2, 16.3-16.3.1
Video:
ISLV: Ch. 8, Video 1-5
Chen, T. (2016) Video: 0-34 minutes		2Mandatory: 57p
4
3Neural network basics, backpropagation (Tensorflow/Keras)DL: 6 (skip 6.3.3, 6.2.2.4, 6.5.2- , 6.6), 7-7.1, 7.4, 7.7-7.8, 7.12, 8-8.2.6 (skip 8.1.3), 8.4, 8.7.1
DLR (optional): 2-2.3, 2.4.4, 3-3.2
Video:
3B1B1: Ch. 1-4		3Mandatory: 105p
Optional: 23p
5
4CNN, fine tuning/transfer learning, hyper parameter searchDL: 9-9.4, 11-11.5, 15.2
DLR: 5.1, (5.2, 5.3 is optional, but good for the assignment)
Video:
Ng (2017)
3B1B2		4Mandatory: 50p
Optional: 24p
6
Deadline: Project description 1
7
5Recurrent Nets, Transformer architectures, BERTDL: 10-10.2 (skip 10.2.1),10.4, 10.10
DLR: 6-6.2
JM: 10-10.2,11-11.3
Alammar, J. (2018b, 2018c)
Optional: Alammar, J. (2018a)
Devlin et al (2018)
Vaswani et al (2017)
Olah (2015)
Video:
Dirac (2019)
Phi (2020)		529+35+8+8+8+2+11+15= 116p
8
II6Introduction to unsupervised learning, pPCA, K-means, Mixture of Gaussians, The EM algorithmESL: 6.8, 8.5-8.5.2, 13 - 13.2.1, 13.2.3, 14-14.1, 14.3 (skip 14.3.9, 14.3.10, 14.3.12), 14.5-14.5.1
DL: 5.8, 13-13.1, 19.2
Smyth (2020)
Ng (2019)
Video:
ISLV: Ch. 10, Video 1-3		62+5+5+2+18+10+2+5+2+1+3=60p
9
7Variational autoencoders, topic models, Large Language ModelsDL: 14-14.1
Kingma and Welling (2019): Ch. 1-2.8
Hand (2020)
Griffiths and Steyvers (2004)
Blei (2012)		732 + 24 + 6= 62p
10
III8Introduction to reinforcement learning, exploitation vs exploration, bandits, finite Markov decision processesSutton and Barto (2020): Ch 1-3 (skip 1.7-1.8, 2.8-2.9, and historical remarks)860p
11
IVProject work, Ethics in AIVerma and Rudin (2018),
Video:
UnHeard (2023)
12
Project work
13
Deadline: Project report
14
Deadline: Second return of assignments
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100