SUBHANKAR MISHRA

ଶୁଭଙ୍କର ମିଶ୍ର

Reader-F, School of Computer Sciences, NISER
ପାଠକ-ଏଫ, ସଂଗଣକ ବିଜ୍ଞାନ ବିଦ୍ୟାଳୟ, ନାଇଜର

CS460/660 - Machine Learning 2023 (2022-23 Even Semester)

Syllabus

Learning, Inductive Bias, Features, Labels, Basics of Statistics and Probability
Supervised Learning - Some models
Ensemble Methods: Bagging, Boosting. Learning Theory
Unsupervised Learning - Some models
Reinforcement Learning Basics (if time permits)

Expected skills (No mandatory prerequisite)

Algorithms (CS301 or equivalent)
Python
Probability and Statistics

Grading scheme

Grading will be absolute.

Project - 40 marks
Assignment - 10 marks
Midterm - 10 marks
Endterm - 40 marks
Total -100 marks

Lectures

Week	Lec #	Topic	Slide
Week 1	1	Course Logistics and Introduction
	2	Generalization	Group1
	3	Why and when Machine Learning,
	4	Inductive Learning, Structured Data
Week 2	5	Decision Trees	Group2
	6	Inductive Bias	Group2
	7	Expected Loss, Inductive vs Transductive learning	Group3
	8	General approach for Machine Learning	Group3
Week 3	9	Types of Machine Learning (Based on Input and Output)	Group4
	10	Supervised Learning - Linear SVM	Group5
	11	Supervised Learning - Linear SVM (Continued)	Group5
	12	Linear Regression	Group6
Week 4	13	Linear Regression (Continued)	Group6
	14	Gradient Descent	Group7
	15	Gradient Descent (Continued)	Group7
	16	Logistic Regression	Group8
Week 5 (Jan 30 - Feb 3)		Project Proposal Presentations
Week 6	-	Quiz - 1
	17	Decision Trees - CART	Group9
	18	Decision Trees - CART - Continued	Group9
	-	Decision Trees - ID3	Group10
	19	SVM - Hinge Loss	Group11
Week 7	20	SVM - Kernel Functions CS229 Notes - Andrew Ng	Group12
	21	SVM - Kernel Functions - Continued	Group12
	22	kNN	Group13
	-	kNN Implementation	Group13
Week 8	23	Feature Engineering (One-hot encoding, bucketing, normalization, standardisation, missing features, imbalanced dataset)	Group14
	24	Model Performance Assessment	Group15
	25	Hyperparameter Tuning	Group15
	-
Week 9	26	Unsupervised Learning - Clustering - kMeans	Group16
	27	Unsupervised Learning - Clustering - DBSCAN
	28	Unsupervised Learning - How to find the number of clusters?
Week 10,11 (Mar 6 - Mar 17)		Project Midterm Presentations
Week 12	29	Unsupervised Learning - Dimensionality reduction PCA	Group17
	30	Unsupervised Learning - Dimensionality reduction tSNE	Group17
	31	Bias and Variance	Group18
	32	Regularization	Group18
Week 13	33	One-class classification and Multiclass classification	Group19
	34	Multilabel classification	Group19
	35	Ensemble Learning - Bagging	Group20
	36	Ensemble Learning - Boosting	Group20
Week 14	37	XGBoost	Group21
	38	AdaBoost	Group21
	39	Naive Bayes Classifier	Group22
	40	Cross Entropy Loss	Group23
Week 15,16 (Apr 10 - Apr 21)		Project Final Presentations

Assignments

[05] Assign 1: Lecture Note Slides

Same as project group, Overleaf template
Cover the content of the class
Add additional examples and sample codes
Follow the github clone and pull request to update your slides (similar to project submissions)
Update: 20 Jan 2023 - Upload both .pdf as well as .zip containing the source for your slides.

[05]

Project

A group of 2. Class project must be something new that you did in this semester. Two types of projects are allowed:

Category - 1: You pick/create an interesting (new) dataset, apply best suited one or more well known machine learning algorithms as baselines and extend these baselines in your creative and interesting way.
Category - 2: A theoretical project should look at an open question, definite it concretely, look at the literature, design and develop creative and interesting attempts.

Deliverables for reports will be in LaTeX generated pdf. Format should follow NeurIPS template. Updates:

25 Jan, 2023 - You need to run a plagiarism check for all the reports to submit and share the result as well.
02 Feb, 2023 - The codes and dataset will be shared over github (private repo is fine if the work is to be published.)

[05] (Max. 2 slides) Project Proposal - 5 mins

Title, dataset, idea, relevant papers, teammate with work division, what to do by Midway, what baselines to implement, what are the expected results

[10] (Max. 10 slides) Project Midway - 15 mins + 5 mins Q&A, Midterm Report - Max Limit 4 pages

Insightful analysis of 2-3 related papers.
Experiments you have done along with results.

[15] (Max. 15 slides) Final Project Presentation - ~~20 mins + 10 mins~~ 10 mins, Endterm Report - Max Limit 8 pages

Review
New and interesting attempts to solve the problem
Plan for further work for submission

[10] Paper Submission
- [5] Average conference
- [10] Top conference - Deadlines
Conference/Journal list: The final selection of the conference will be in coordination with the instructor.

Submission guidelines:

Clone GitHub
index.html

If this is the first time, create a new row in the table, add your members, project title and initialize the groups slides and reports.
You will add/update the links associated with your group.

Projects folder

If this is the first time, create a new folder under projects folder
Add/update your files in that folder only. Do not make any changes other than your group/individual folder

Create a pull request after the changes are done

#	Team Members	Title	Proposal	Midterm	Final	Paper
1	Sagar Prakash Barad	Estimation of Electronic Band Gap Energy Using Machine Learning	Slide	Slide	Slide	Slide
1	Sajag Kumar		---	Report	Report	Report
2	Aritra Mukhopadhyay	Improvement on the Quaternion-based models: extension to larger datasets and Batch Normalization	Slide	Slide	Slide	Slide
2	Adhilsha A		---	Report	Report	Report
3	Aniket Nath	Using Self-supervised pre-trained model fine-tuned with Active Learning algorithm to find Dual Active Galactic Nuclei in SDSS Dataset.	Slide	Slide	Slide	Slide
3	Diptarko Choudhury		---	Report	Report	Report
4	Debashish Paik	Analysis of Economy using nighttime light data	Slide	Slide	Slide	Slide
4	Rohan Naskar	Analysis of Economy using nighttime light data	---	Report	Report	Report
5	Dev Shrivastva	Detection of Retinal Diseases	Slide	Slide	Slide	Slide
5	Shubhanshu Prasad	Detection of Retinal Diseases	---	Report	Report	Report
6	Priyanshu Parida	Differentiation of Neutron and Gamma Response for EJ-301 Detector at low energy	Slide	Slide	Slide	Slide
6	Nehal Khosla		---	Report	Report	Report
7	Arshia Anjum	Infer accreted mass fractions of central galaxies	Slide	Slide	Slide	Slide
7	Sibabrata Biswal	Infer accreted mass fractions of central galaxies	---	Report	Report	Report
8	Dibya Bharati Pradhan	Exoplanetary surface composition prediction using ML	Slide	Slide	Slide	Slide
8	Oommen P Jose	Exoplanetary surface composition prediction using ML	---	Report	Report	Report
9	Ayush Singhal	Analysing Exoplanetary Atmospheres using ML	Slide	Slide	Slide	Slide
9	Gaurav Shukla	Analysing Exoplanetary Atmospheres using ML	---	Report	Report	Report
10	Akankshya Nayak	Expressivity of geometric GNN	Slide	Slide	Slide	Slide
10	Soumya Dasgupta	Expressivity of geometric GNN	---	Report	Report	Report
11	Chandan Kumar Sahu	Studying the interior evolution of rocky exoplanets using machine learning	Slide	Slide	Slide	Slide
11	Maitrey Sharma		---	Report	Report	Report
12	Anna Binoy	Application of ML in Predicting Gaseous Properties of Earth's Atmosphere	Slide	Slide	Slide	Slide
12	Sumegha M.T.		---	Report	Report	Report
13	Rahul Vishwakarma	Indoor localization using WiFi RSSI fingerprinting	Slide	Slide	Slide	Slide
13	Jyotish Kumar J.	Indoor localization using WiFi RSSI fingerprinting	---	Report	Report	Report
14	Arpan Maity	Gradient Boosted Decision Trees and their application in improvising identification of decay of Higgs Bosons into pair of electrons	Slide	Slide	Slide	Slide
14	Krishna Kant Parida		---	Report	Report	Report
15	Anuleho	Using Federated Transfer Learning to correlate sleeping health to digital wellbeing	Slide	Slide	Slide	Slide
15	Abhijit		---	Report	Report	Report
16	Ratul Das	Phase Transition Detection without Order Parameters	Slide	Slide	Slide	Slide
16	Mihir Chandra	Phase Transition Detection without Order Parameters	---	Report	Report	Report
17	Ritadip Bharati	ML in Fashion	Slide	Slide	Slide	Slide
17	Sudip Kumar Kar	ML in Fashion	---	Report	Report	Report
18	Krishanu Kanta	Analysing Seismic data of Palghar using ML.	Slide	Slide	Slide	Slide
18	Sagnik Rout	Analysing Seismic data of Palghar using ML.	---	Report	Report	Report
19	Sunaina	Predicting breeding sites of Desert Locust through environmental parameters	Slide	Slide	Slide	Slide
19	Bibhu		---	Report	Report	Report
20	Fida Salim	Retrieving Pressure-Temperature and Water Vapour Profiles in Earth’s Atmosphere from INSAT 3DR data using Machine Learning	Slide	Slide	Slide	Slide
20	Soumik Bhattacharyya		---	Report	Report	Report
21	Deependra Singh	Retinal Fundus Multi-Disease Image Classification	Slide	Slide	Slide	Slide
21	Saksham Agarwal	Retinal Fundus Multi-Disease Image Classification	---	Report	Report	Report
22	Kaling Vikram Singh	Prediction of Lattice Structure of perovskites from physical properties using ML	Slide	Slide	Slide	Slide
22	Sunil S Raja		---	Report	Report	Report
22	Summit Bikram Nayak	Quantitative Analysis of Lipid Droplets using Image processing	Slide	Slide	Slide	Slide
22	Ankur Abhijeet		---	Report	Report	Report

Books

Some recommended books on Machine learning.

The Elements of Statistical Learning Link
Mathematics for Machine Learning Link
Introduction to Machine Learning with Python Link
A course in Machine Learning Link

Academic Integrity

Any plagiarism, copying, allowing copying, unpermitted aid will lead to 'zero' in the assignment/exam/project.