Syllabus#
Welcome to STATS320!
This course is cross-listed 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: Noah Cowan
Spring Quarter, 2024-25
Stanford University
Course Description#
With modern high-density electrodes and optical imaging techniques, neuroscientists routinely measure the activity of hundreds, if not thousands, of cells simultaneously. Coupled with high-resolution 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 high-dimensional neural and behavioral time-series; 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 in-class 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: MW 1:30pm-2:20pm
Location: MW in STLC115
Grading: Credit or letter grade
Components: Lectures on Mon/Wed
Office Hours:
Scott: 10:30am-12:00pm Tuesday, Wu Tsai Neurosciences Institute, 2nd floor by the NeuroTheory center
Noah: 10am-11:30am Monday, CoDa room B06
This course will have a final project, not an exam
Book#
We will use an online textbook that I have been developing over the past few years called Machine Learning Methods for Neural Data Analysis. It’s a work in progress, and I will continue to update it throughout the quarter!
Schedule#
The lectures develop the theory behind the methods developed in the labs (i.e., homework assignments). I’ve organized the course into four units: signal extraction, encoding and decoding, unsupervised modeling, and current research. 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 |
Neuro Topic |
ML Topic |
Reading |
Assignment |
|---|---|---|---|---|
Mon, Mar 31 |
Course Overview |
|||
Wed, Apr 2 |
Spike Sorting |
Mixture Models |
Lab 1 Out |
|
Mon, Apr 7 |
Calcium Deconvolution |
Matrix Factorization |
||
Wed, Apr 9 |
Calcium Deconvolution |
Convex Optimization |
Lab 1 Due |
|
Mon, Apr 14 |
Markerless Pose Tracking |
Convolution and Cross-Correlation |
||
Wed, Apr 16 |
Markerless Pose Tracking |
Convolutional Neural Networks (CNNs) |
Lab 2 Due |
Unit II: Encoding and Decoding Models for Neural Data#
Date |
Neuro Topic |
ML Topic |
Reading |
Assignment |
|---|---|---|---|---|
Mon, Apr 21 |
Neural Encoding |
|||
Wed, Apr 23 |
Neural Encoding |
Lab 3 Due |
||
Mon, Apr 28 |
Bayesian Decoding |
Bayesian Inference |
||
Wed, Apr 30 |
Bayesian Decoding |
Markov Chain Monte Carlo (MCMC) |
Lab 4 Due |
Unit III: Unsupervised models of neural and behavioral data#
Date |
Neuro Topic |
ML Topic |
Reading |
Assignment |
|---|---|---|---|---|
Mon, May 5 |
Behavioral Segmentation |
Hidden Markov Models (HMMs) |
||
Wed, May 7 |
Behavioral Segmentation |
Expectation Maximization (EM) |
Lab 5 Due |
|
Mon, May 12 |
Neural Dynamics |
Linear Dynamical Systems (LDS) |
||
Wed, May 14 |
Neural Dynamics |
Switching Linear Dynamical Systems (SLDS) |
Lab 6 Due |
Unit IV: Current Research Topics#
Date |
Topic |
Reading |
Assignment |
|---|---|---|---|
Mon, May 19 |
(Sequential) Variational Autoencoders |
||
Wed, May 21 |
Gaussian Process State Space Models |
Lab 7 Due |
|
Mon, May 26 |
Memorial Day, No Class |
||
Wed, May 28 |
(Neuromodulated) Recurrent Neural Networks |
||
Mon, Jun 2 |
Deep State Space Models |
||
Wed, Jun 4 |
Foundation Models for Neuroscience |
||
Mon, Jun 9 |
Final Project Due |
Labs#
Each week, you will implement a minimal version of the method we discussed in lecture. These labs will be your assignments.
You must work in a team of two people.
There’s a catch! You may not work with the same person twice. (We will have a discussion forum on Ed to facilitate matching.)
Lab reports will be submitted via GradeScope.
All assignments are due at 11:59pm PT on the specified date.
Final project#
You will work on the final project in teams of 2 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 TBD.
The final report will be due Mon, June 9 at 11:59pm.
Late Policy#
We will allow 7 late days to be used as needed throughout the quarter. Since assignments are done in teams of two, both students must have sufficient late days to turn in a late assignment. Unfortunately, we cannot allow late days on the final project.
Grading#
7 labs: 10% each, total 70%
Final project: 25%
Participation: 5%
Note: You must do a final project in order to pass.