ABCDEFGHIJKLMNOPQRSTUVWXYZ
1
WeekDateTopic Paper 1Paper 2P1 advocateP1 criticP2 advocateP2 criticOPTIONAL
2
18/22class introduction and overview
3
18/24class topic overview: Tour de SysML 2021MLSys
4
18/26registration deadline
5
28/29Big ideas from ML and Systems
Berkeley View of Systems Challenges for AI
CACM12
6
28/31ML LifecycleTFX, KDD17Hidden technical debt in ML
7
39/5LABOR DAY (no class)
8
39/7ML SystemsTheanoAutoDiff in ML
9
49/12Declarative ML Frameworks and AutoDiffTensorflow
10
49/14Imperative ML Frameworks & Project IdeasPytorchTensorflow Eager
11
59/19General ML Frameworks (Ray)AlpaRay
12
59/21Prediction Serving: Single ModelClockworkClipper
13
69/26Prediction Serving: Model compositionInferline
14
69/28Prediction Serving: Model ZooINFaaS
15
710/3Hyperparameter OptimizationOptunaRubberbandHyperSched
16
710/5HyperOpt continued + NAS OverviewHW-NAS Surveydarts-iclr19NAS SurveyMNasNet-cvpr19
mobilenetv3-iccv19
proxylessnas-iclr19
17
810/10Neural Architecture Search: Latency-awarenessOFA-iclr20bignas-eccv20EfficientNet
18
810/12guest lecture: Mixed Precision DNNPWLQ-eccv20MixedPrecDNN-mlsys21
19
910/17FALL BREAK (no class)
20
910/19Resource Management for MLCoDDLGavel-OSDI2020TetriSched
21
1010/24Resource Management for MLESCHER-socc22Synergy-OSDI22
22
1010/26mid-point project presentations
23
1110/31mid-point project presentations
24
1111/2mid-point project presentations
25
1211/7guest lectureFlexFlow@MLSys19BlueConnect
26
1211/9guest lectureEFLOPS@HPCA20ISCA20
27
1311/14Serverless Frameworks for ML
Serverless Computing: One Step Forward, Two Steps Back
FaasML
28
1311/16Federated LearningFL@GoogleFL@Scale@Google
29
1411/21AI for healthcare: SysML opportunitiesHOLMES-kdd20UnfoldML-neurips22DL4EHR SurveyDL4ECG
30
1411/23STUDENT RECESS (no class)
31
1511/28final project presentations
32
1511/30final project presentations
33
1612/5final project presentations
34
35
36
10/29withdrawal deadline
37
12/9final project video/demo due by 2359ET
38
12/10final project report due by 2359ET
39
40
41
NOTESfinal instructional days: Dec 5-6
42
final exam slot: (N/A)
43
44
45
Bonus Topics
ML CompilersTVM
46
ML FrameworksMxnet
47
Model CompressionSqueezenetRethink net pruningshufflenet
48
Reinforcement Learning (Brief Intro) + Concluding Remarks
Distrib RL ICLR18
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