The goal of my project with Dr. Balasubramanian was to utilize modern machine learning techniques, especially deep learning, to learn more about the visual processing pathway within the brain. Over the course of the summer, I was granted the opportunity to work closely with another student with a PURM grant, Jordan Lei, as well as other Post-Docs within the lab. We worked with neural firings collected both experimentally from retinal neurons and computationally from a programmed model retina we created. These firings were then used to train deep neural networks, a form of machine learning, to solve classification problems so that we can analyze the neural network to gain insight into the human brain. The end goal would be to create a network and model retina that can cluster together objects scene without any labels or supervision to ‘learn’ the world presented to our model neurons.
Over the course of this project, I was able to gain a lot of exposure to areas I had not touched before. I learned about some of the basic mathematical principles within computational neuroscience research. I also gained exposure to basic implementations of several machine learning algorithms as well as extensive exposure to implementations of neural networks. I also was able to gain technical skills, such as the ability to program in Matlab as well as increased proficiency in Python and familiarity with APIs such as TensorFlow, matplotlib, and NumPy. These skills in programming will probably be very useful in my future academic ventures, as many researchers utilize these two languages in their work. I am also hoping to be able to apply these machine learning techniques to programming projects of my own in the future. I also was able to learn how to read math heavy research papers and more efficiently read through mathematical notation.
The goal of my project with Dr. Balasubramanian was to utilize modern machine learning techniques, especially deep learning, to learn more about the visual processing pathway within the brain. Over the course of the summer, I was granted the opportunity to work closely with another student with a PURM grant, Jordan Lei, as well as other Post-Docs within the lab. We worked with neural firings collected both experimentally from retinal neurons and computationally from a programmed model retina we created. These firings were then used to train deep neural networks, a form of machine learning, to solve classification problems so that we can analyze the neural network to gain insight into the human brain. The end goal would be to create a network and model retina that can cluster together objects scene without any labels or supervision to ‘learn’ the world presented to our model neurons.
Over the course of this project, I was able to gain a lot of exposure to areas I had not touched before. I learned about some of the basic mathematical principles within computational neuroscience research. I also gained exposure to basic implementations of several machine learning algorithms as well as extensive exposure to implementations of neural networks. I also was able to gain technical skills, such as the ability to program in Matlab as well as increased proficiency in Python and familiarity with APIs such as TensorFlow, matplotlib, and NumPy. These skills in programming will probably be very useful in my future academic ventures, as many researchers utilize these two languages in their work. I am also hoping to be able to apply these machine learning techniques to programming projects of my own in the future. I also was able to learn how to read math heavy research papers and more efficiently read through mathematical notation.