Hack University syllabus with attendance
 Share
The version of the browser you are using is no longer supported. Please upgrade to a supported browser.Dismiss

 
View only
 
 
ABCDEFGHIJKLMNOPQRSTUVWYZAAABACADAEAFAGAHAIAJAKALAM
1
Hack University Machine Learning
Grades, attendance, notes, etc. For the ML Course, "attendance" or "grade" will be a fraction between 0 and 1
2
DateTeacherTATopicSubtopicSubtopicSubtopicLecturePythonStudent 1Student 2Student 3Student 4Student 5Student 6Student 7Student 8
3
day12/11ThuHobsGetting StartedApplications and types of problems, ethics, machine learning morals, the control problemModeling, forecasting, inference, correlation, experimentingEnvironment setup, python, linux, virtualbox, vagrant, ssh, matplotlib, etc/docs/day1/huml/day1
4
day22/14SunHobsBest-Practices software developmentGit and GitHubMore environment setupBot examples: Will, Slackbot, twip
API examples: hack oregon plotpdx project, twitter api, slackapi
Python tools for apis: requests, pyton-rtmbot, twython, boto
/docs/day2/huml/day2
5
day32/18ThuHobsDanBots and Web CrawlersDIY Web CrawlerGraph Theory, Graph Data Structures, Regular Expressions, FSA, NFA, DFA, Data-driven video games vs choose-your-own adventure games/booksDIY Chat Bots (twitter or slack)/docs/day3/huml/day3
6
day42/21SunHobsHobsNatural Language ProceesingBag of words, word vectors, word statistics, n-grams"Computing" words, "information" metrics, redundancy, word vectors, Dimension Reduction, Latent Dierlicht Allocationtwip, semantic indexing of text files on your laptop/docs/day4/huml/day4
7
day52/25ThuEbrahimTime SeriesWorking with time series dataDimension Reduction, PCA vs LDA, scatter matrices, saddle point proliferation in high dim spacesFinancial and weather (chaotic) time series, event studies, autocorrelation, allan variance/docs/day5/huml/day5
8
day62/28SunHobsNLPNLTKpython-rtmbotTFIDF, PCA/docs/day6/huml/day6
9
day73/3ThuHannesHobsNeural NetsPerceptronsMultilayer PerceptronsHyperparameter Optimization? Tools?/docs/day7/huml/day7
10
day83/6SunColeHobsCNNsConvolutional Neural NetsDeep CNNs, regularization, random dropout, feature engineeringRepresentation visualization (hidden layers)? Tools?/docs/day8/huml/day8
11
day93/10ThuZekeHobsBig OAnalyzing algorithms for CPU and RAM optimization. Data structures and algorithm patterns. The "algebra" of Big ON-p hard problems, quantum computers, scalability, parallelizable algorithms, GPUs, limitations and bottle necks, hadoop, gensim, theanoComputational Linguistics (natural language statistics, grammar). Big O for TFIDF/docs/day9/huml/day9
12
day103/13SunHobsProject Day and Python DoctestsNLP, python-rtmbot, Python doctestsPython functionsgensim.summarize/docs/day10/huml/day10
13
day113/17ThuGrimmHobsClassificationDecision trees: Classification, supervised learning, information theory.Random forests: ensemble learning.Kmeans and unsupervised learning -or- Project: Revisit Crime Data/docs/day11/huml/day11
Problems: Supervised v unsupervised v reinforcemnet, classification v regression.
Methods: Bayesian vs Decision trees, Random Forests, Multivariate.
Performance: variance and bias, RMSE, outliers.
Project: examine the Crime Data in more detail
14
day123/20SunHobsProject Day and Word2Vec demossh and tmux on totalgood.org to watch server provisioning for gensim.Word2Vecgutenberg collection sentence segmentation and Word2Vec trainingloading pre-trained Word2Vec model (GoogleNews 300D)/docs/day12/huml/day12
15
day133/24ThuHobscontrol theoryfeedback control, pid, stability (Tacoma Narrows bridge), fundamental frequency, resonance, psd, transfer functionDSP, discrete-time filters, convolution, kalman filters, system identification, adaptive control, optimal control, low pass filter, oatftrjectory planning, receding horizon, path planning, geospatial data and problems, transformations, rotations, vector spaces/docs/day13/huml/day13
16
day143/27SunHobsProjects (NLP for forum spam filtering, web scraping titles, home price prediction) Distances (vector norms) in high dimensional space (TFIDF vectors, word vectors)Clustering in high dimensional space/docs/day14/huml/day14
17
day153/31ThuGrimm? Cole? Hannes?AIturning test, measures of complexity, measures of intelligence, measures of information and entropy, predictabilityinference, training, logic, ontologies, freebase, CYCstate of the art, latest research, FFLabs, NIPS, trends, control problem, what's next/docs/day15/huml/day15
18
day164/3SunHobsHack Oregon projectsprediction and ML problems, visualization problemsSidenotes. AI.The control problem. Cooperating machines. Competitive agents. Genetic algorithms. Training/reinforcement for ethics, moral behavior./docs/day16/huml/day16
19
20
21
22
23
Skipped Lectures
24
graph search, graph theorytree vs graph, directed graph, dependency trees, scale free graph, fully connected graph, bipartite, depth first, breadth first, dykstra, astarautomatic heuristic formulation
25
?HobsNatural Language Proceesing"Computing" words, "information" metrics, redundancy, word vectorsDimension Reduction, Latent Dierlicht Allocationtwip, semantic indexing of text files on your laptop
26
Binary Classification, Multiclass Classification, Clustering
27
DataPreprocessing, cleaning, validatingDatabases, data structures,
28
statistical modelsStandard deviation, RMSE, bias vs variance tradeoff (Hannes), Bayes rule, bayes classifier, naive bayes classifierhidden markov model, markov chain, decision treesneural nets to predict weather
29
image processingfeature generation, representation vs compression, optical flow, stereoscopic processing, structured light, 3d reconstruction, video processing, compressiontracking, segmentation/docs/day12/huml/day12
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
3. Regression
58
- Linear Regression
59
- Nonlinear Regression
60
- Multivariate Regression
61
4. Information Theory
62
- Entropy
63
- Thermodynamics
64
- Information
65
- Black Holes and an Expanding Universe
66
5. Supervised Classification
67
- Application/Motivation
68
- Logistic Regression
69
- Derived from Linear Regression
70
- Decision Tree
71
- Random Forest
72
- Neural Nets
73
- SVM
74
- Genetic Algorithms useful when
75
- fit-test is fast
76
- thoughtful design is hard
77
6. Unsupervised Classification
78
- Applications/Motivation
79
- Clustering
80
- KMeans
81
- KNN
82
- Neural Nets
83
- Word2Vec
84
7. Ensemble Methods
85
8. Meta-parameter search (tuning your learner)
86
- Grid Search
87
- Random Search
88
- Graph search with heuristic
89
9. Supervised v Unsupervised
90
- Labels
91
- Imbalanced Data Sets
92
10. Bayesian Inference
93
- Naive Bayes
94
- Non-naive bayes (Markov Models)
95
11. More Neural Nets
96
- Convolutional
97
- Recurrent
98
- Recursive
99
- Deep
100
- Examples: image processing
Loading...
 
 
 
Machine Learning
Data Engineering
ML Applications