Introduction#
Welcome to STATS320!
This course is crosslisted as STATS220, NBIO220 and CS339N. They are all the same. Enroll in the version that is best for your degree requirements.
Instructor: Scott Linderman
TAs: Sifan Liu and Ying Jin
Winter Quarter, 202223
Stanford University
Course Description#
With modern highdensity electrodes and optical imaging techniques, neuroscientists routinely measure the activity of hundreds, if not thousands, of cells simultaneously. Coupled with highresolution behavioral measurements, these datasets offer unprecedented opportunities to learn how neural circuits function. This course will study statistical machine learning methods for analysing such datasets, including: spike sorting and calcium deconvolution techniques for extracting relevant signals from raw data; markerless tracking methods for estimating animal pose in behavioral videos; state space models for analysis of highdimensional neural and behavioral timeseries; point process models of neural spike trains; and deep learning methods for neural encoding and decoding. We will develop the theory behind these models and algorithms and then apply them to real datasets in the inclass coding labs and final project.
Prerequisites#
You should be comfortable with basic probability (STATS 116) as well as multivariate calculus and linear algebra. This course will emphasize implementing models and algorithms in Python, so coding proficiency is important. We will have a coding primer in the first week to help get you up to speed if you’re coming from R or Matlab.
Logistics#
Time: MWF 11:30am12:20pm
Location: MW in 380380Y, F in Turing Auditorium (Polya Hall)
Grading: Credit or letter grade
Components: Lectures on Mon/Wed, inclass labs on Fri
Office Hours:
Sifan: 6:007:30pm Tues, Sequoia Hall, Room 207
Scott: 1:302:30pm Wed, Wu Tsai Neurosciences Institute, 2nd floor by the NeuroTheory center
Ying: 11:0012:30pm Thurs, Sequoia Hall, Room 207
This course will have a final project, not an exam
Schedule#
The lectures develop the theory behind the methods developed in lab. I’ve organized the course into three units: signal extraction, encoding and decoding, and unsupervised modeling. At the end, you’ll work on a final project in which you will use, explore, or extend the techniques studied in class.
Unit I: Extracting biological signals from raw data#
Date 
Type 
Topic 

Mon, Jan 9 
Lecture 1 
Course overview 
Wed, Jan 11 
Lecture 2 

Fri, Jan 13 
Lecture 3 
Basic neurobiology 
Mon, Jan 16 
MLK Day 
No class 
Wed, Jan 18 
Lecture 4 

Fri, Jan 20 
Lab 1 

Mon, Jan 23 
Lecture 5 

Wed, Jan 25 
Lecture 6 

Fri, Jan 27 
Lab 2 

Mon, Jan 30 
Lecture 7 

Wed, Feb 1 
Lecture 8 

Fri, Feb 3 
Lab 3 
Unit II: Encoding and decoding models for neural data#
Date 
Type 
Topic 

Mon, Feb 6 
Lecture 9 

Wed, Feb 8 
Lecture 10 

Fri, Feb 10 
Lab 4 

Mon, Feb 13 
Lecture 11 

Wed, Feb 15 
Lecture 12 

Fri, Feb 17 
Lab 5 
Unit III: Unsupervised models of neural and behavioral data#
Date 
Type 
Topic 

Mon, Feb 20 
Pres. Day 
No class 
Wed, Feb 22 
Lecture 13 
Mixture Models, EM and Hidden Markov models (HMMs) 
Fri, Feb 24 
Lab 6 

Mon, Feb 27 
Lecture 14 
More HMMs 
Wed, Mar 1 
Lecture 15 
Switching linear dynamical systems (SLDS) 
Fri, Mar 3 
Lab 7 

Mon, Mar 6 
Lecture 16 
Variational Autoencoders (VAEs) 
Wed, Mar 8 
Lecture 17 
Sequential VAEs 
Fri, Mar 10 
Lab 8 

Mon, Mar 13 
Guest Lec. 

Wed, Mar 15 
Lecture 19 
Current research 
Fri, Mar 17 
Presentations 
Project presentations 11:201:20pm, with lunch provided 
Labs#
You will work start the labs in class on Fridays, so attendance is required. (If you are sick or have a onetime conflict, please let me know as soon as possible.)
You will be automatically assigned to a team of 3 students.
The labs will probably take more than one class period to complete. You should find time to work with your teammates outside of class to finish them.
The labs are due the following Thursday night at 11:59pm.
Lab reports will be submitted via GradeScope.
Final project#
You will work on the final project in teams of 23 people (you choose your team!)
You must use real neural or behavioral data. We will provide links to suggested datasets, or if you are an experimentalist, you can use your own!
A project proposal will be due Monday, Feb 27 at 11:30am.
The final report will be due Friday, Mar 24 at 11:30am.
Grading#
First 6 labs: 10% each, total 60%
Final project presentation: 10%
Final project report: 20%
Participation: 5%
Note: We will work on Labs 7 and 8 in class, but they will not be graded. That way you can focus on final proejcts in March. Your attendance on those days will count toward the class participation grade.