Neural Signal Processing & Machine Learning

Why do "Neural Signal Processing"?

Why don't I steal a quote from the original course website?

The activity of a complex network of billions of interconnected neurons underlies our ability to sense, represent and store the details of experienced life, and enables us to interact with our environment and other organisms. Modern neuroscience techniques enable us to access this activity, and thus to begin to understand the processes whereby individual neurons work together to enable complex behaviors. In order to increase this understanding and to design biomedical systems which might therapeutically interact with neural circuits, advanced statistical signal processing and machine learning approaches are required.

What does this course cover?

Why don't I steal another quote from the original course website?

This class will cover a range of techniques and their application to basic neuroscience and neural interfaces. Topics include an introduction to neurobiology and electrophysiology for engineers, point processes, dimensionality reduction, classification/clustering, spectral analysis, and genetic and optical tools for interrogating neural circuits. Neuroscience applications include modeling action potentials and firing rates, automated analysis of activity-dependent fluorescence imaging data, decoding, spike sorting, and field potential analysis. This course is open to students with no prior neurobiology coursework.

Shay's teaching/grading philosophy

I personally believe every student who wants to learn and meets the prerequisite knowledge can indeed learn all of the material. However, grades and hard deadlines in general take away from the learning opportunity and promote a culture of trying to get the correct answer in the minimum amount of time possible without encouraging creativity and allowing students to make mistakes in order to really learn. As a result, each homework assignment will have expected outcomes and objectives listed with students being requested to explain in a sentence or two how the answers or plots demonstrate those expectations. If we, the course staff, are in agreement, full credit shall be given; otherwise, we will provide feedback for reworking the problem for full credit. Yes, full credit. If someone makes a mistake in life or academics and they are judged/evaluated and punished in any sense (numerically, here) without an opportunity to correct their mistake, how does that foster learning? Mistakes with the opportunity to correct mistakes based on feedback is the only way any of us can truly be brave enough to try new things, explore variety of solutions and truly learn! Who knows, there may be solutions and explorations of the data in this course that we, the course staff, haven't even seen or thought about before! Lastly, all deadlines given will be soft. This should to combat the need to finish quickly allowing for robust exploration of the problem and fully meeting each of the expected objectives/outcomes. Additionally, stress levels should be lower and promote mental wellbeing given that students do not need to stay up all night in order to finish homeworks. To enable functionality of this process and make sure assignments are completed such that the course can be completed within allotted semester timeline, students must contact me and explain why they are unable to meet the soft deadline --- especially if they are struggling with the assignment as I would love to help! Learning is a team effort. Let's do this together! We, the course staff, are here to HELP TEACH not judge and evaluate.