ECE228 and SIO209 Machine learning for physical applications, Spring 2018

Final Projects

Group	Topic	Authors	Poster	Report
1	Reimplementation of source localization in an ocean waveguide using supervised learning	Jinzhao Feng, Zhuoxi Zeng, Yu Zhang	Poster	Paper
2	Machine learning methods for ship detection in satelite images	Yifan Li, Huadong Zhang, Xiaoshi Li, Quianfeng Guo	Poster	Paper
3	Transparent Conductor Prediction	Yan Sun, Yiyuan Xing, Xufan Xiong, Tianduo Hao	Poster	Paper
4	Ship identification in sateklite Images	Weilun Zhang, Zhaoliang Zheng, Mingchen Mao,	Poster	Paper
5	Fruit Recognition	Eskil Jarslkog, Richard Wang, Joel Andersson	Poster	Paper
6	RSNA Bone Age Prediction	Juan Camilo Castillo, Yitian Tong, Jiyang Zhao, Fengcan Zhu	Poster	Paper
7	Facial Expression Classification into Emotions	David Orozco, Christopher Lee, Yevgeniy Arabadzhi, Deval Gupta	Poster	Paper
8	Urban Scene Segmentation for Autonomous Vehicles	Hsiao-Chen Huang, Eddie Tseng, Ping-Chun Chiang, Chih-Yen Lin	Poster	Paper
9	Face Detection Using Deep Learning	Yu Shen, Kuan-Wei Chen, Yizhou Hao, Min Hsuan Wu	Poster	Paper
10	Understanding the Amazon Rainforest using Neural Networks	Naveen Dharshana Ketagoda, Christian Jonathan Koguchi, Niral Lalit Pathak, Samuel Sunarjo	Poster	Paper
11	Mercedes-Benz Bench Test Time Estimation	Lanjihong Ma, Kexiong Wu, Bo Xiao, Zihang Yu	Poster	Paper
12	Vegetation Classification in Hyperspectral Image	Osman Cihan Kilinc, Kazim Ergun, Yuming Qiao, Fengjunyan Li	Poster	Paper
13	Threat Detection Using AlexNet on TSA scans	Amartya Bhattacharyya, Christine H Lind, Rahul Shirpurkar	Poster	Paper
14	Flagellates Classification via Transfer Learning	Eric Ho, Brian Henriquez, Jeffrey Yeung	Poster	Paper
15	Biomedical Image Segmentation	Lucas Tindall, Amir Persekian, Max Jiao	Poster	Paper
16	“Deep Fakes” using Generative Adversarial Networks (GAN)	Tianxiang Shen, Ruixian Liu, Ju Bai, Zheng Li	Poster	Paper
17	Dog Breed Classification via Convolutional Neural Network	Yizhou Chen; Xiaotong Chen; Xuanzhen Xu	Poster	Paper
18	Dog Breed Identification	Wenting Shi, Jiaquan Chen, Fangyu Liu, Muyun Liu	Poster	Paper
19	Impact of Skewed Distributions on an Automated Plankton Classifier	Will Chapman, Emal Fatima, William Jenkins, Steven Tien, Shawheen Tosifian	Poster	Paper
20	Blood Cell Detection using Single shot MultiBox Detector	Inyoung Huh	Poster	Paper

Professor Peter Gerstoft, Spiess Hall 462, Gerstoft@ucsd.edu
TA Mark Wagner, Spiess Hall, m2wagner@eng.ucsd.edu
TA Paolo Gabriel, Jacobs Hall, pgabriel@eng.ucsd.edu
TA Nima Mirzaee nmirzaee@eng.ucsd.edu
Location: CENTR 109
Time: Monday and Wednesday 5-6:20pm

Sylabus: Machine learning has received enormous interest. To learn from data we use probability theory, which has been the mainstay of statistics and engineering for centuries. The class will focus on implementations for physical problems. Topics: Gaussian probabilities, linear models for regression, linear models for classification, neural networks, kernel methods, support vector machines, graphical models, mixture models, sampling methods, sequential estimation. Prerequisites: graduate standing.

It is not a computer science class, so we go slowly through the fundamentals to appreciate the methods, implement these. We will discuss their use in Physical Sciences. In the first part of the class we focus on theory and implementations. We will then transition to focus on machine learning final projects.

Schedule

Lecture	Date	Topic	Required reading	Assignments
1	April 2	Introduction, probability theory [Slides]	Bishop CH 1.0-1.2
2	April 4	Gaussian probability theory [Slides, Annotated Slides ]	Bishop CH 1.5, 1.6, 2.3
3	April 9	Non Parametric methods, linear models for Regression [Slides Annotated Slides]	Bishop CH 2.5, 3.1, 3.2	Homework 1 due before class
4	April 11	Linear models for Regression [Slides Annotated Slides]	Bishop CH 3.0-3.3	Homework 2 due before class
5	April 16	Sparse models I, Lecture by Santosh Nannuru [Slides Annotated Slides] Paolo Gabriel (TA): Neural brain control. After class we install Tensorflow, Python, Jupyter notebook	Read Chapter 2 for an introduction to sparse problems Murphy 13.1, 13.3, 13.6.1	Homework 3 due before class
6	April 18	Sparse models II, Lecture by Santosh Nannuru [Slides Annotated Slides] Emma Ozanich: Tracking ships using acoustics. Based on our paper: Niu et al, 2017	Murphy 13.7, 13.8	Homework 1 and 2 due 4/20
7	April 23	Linear models for Classification [Slides Annotated Slides]	Bishop CH 4.0-4.3.2	Homework 4 due before class
8	April 25	Backpropagation [Slides Annotated Slides] Eric Orenstein: Automatic Analysis of Planktonic Image Data Sets	Bishop CH 4.3.4, 5.0-5.2
9	April 30	Kernel methods [Slides Annotated Slides]	Bishop CH 6 and from last lecture 5.3 to 5.3.2	Homework 5 due before class
10	May 2	Support vector machines [Slides Annotated Slides]	Bishop CH 7	Homework 6 due before class solution
11	May 7	K-means, EM and Mixture models[Slides Annotated Slides]	Bishop CH 9	Project proposals due
12	May 9	Mike Bianco: Dictionary leaning Trees	Read Bianco's Dictionary learning paper	Homework 7 due before class Solution
13	May 14	Trees, Random Forrest, Gradient boosting> [Slides Annotated Slides] Bayes net	Hastie CH 9 Bishop CH 8.1, 8.2
14	May 16	Graphs, maybe PCA [Slides Annotated Slides]	Bishop CH 8.2 12	Homework 8 due before class
15	May 21	mid-project presentation of final-projects
16	May 23	Sequential estimation[Slides Annotated Slides] Jiuyuan Nie and Chongze Hu "Predict and Design Novel Materials using Machine Learning: in Respect of Experiment and Theory". In prepapration, Read the following paper	Bishop CH 13	Homework 9 due before class
17	May 30	presentation of final-projects II
18	June 4	Introduction to Machine learning Nima Mirzaee: Data assimilation
19	June 6	Final project poster presentation, Jacobs Hall Lobby, 5 pm		Final report due Saturday 16 June.

Books: Main book: Chris M Bishop Pattern Recognition and Machine Learning . A third party Matlab implementation of many of the algorithms in the book.
Other good books:
Hastie and Tibshirani The Elements of Statistical Learning
Kevin P. Murphy: Machine Learning: A Probabilistic Perspective. UCSD license . Matlab codes used in Murphy's book.

Online resources: While not required, I recommend taking these classes. Both are online classes are excellent.
Statistical Learning by Hastie and Tibshirani. My favorite class.
Andrew Ng's Coursera class, Machine learning. This was the first class offered by Coursera.

Grading: Full scale of the letter grade. Grade consists of About 50 % homework and 50% final-project. Your and my purpose is to lean, so a good effort is sufficient.
Homework Cody homework will be graded. I'm a strong believer in leaning by doing. Thus we will have computer-based homework each week.

Final project: Propose a topic before May 1. Otherwise it will be based on my paper: Niu et al, 2017 on arXiv. We will make teams on April 26. Report due June 16.
Suggested topics:

1. Tracking ships using acoustics. Based on my paper: Niu et al, 2017 on arXiv. This can be solved in TensorFlow or SciKit-learn (in Python) or matlab . Data and SVM example
2. Graph signal processing for localizing small events. Based on my paper Riahi 2017.
3. Classifying plankton. This would be based on Jules Jaffe's underwater microscope. This might require convolutional networks, but random forest and support vector machines would also work.
4. Identifying earthquakes from data on a mobile phone (example is in ipython). Extensive example with 3 Gb of data. See background paper in Science

Spring 2017 class