Robot Dreams—IEEE’s Machine Learning Workshop
IEEE SBM organised a two-day Machine Learning certified workshop with an aim to impart adequate technical expertise to students regardless of their prior knowledge of machine learning. The event took place on the 4th and 5th of August, 2018, and witnessed an overwhelming participation of over 50 students from MIT, and 10 outstation students from Canara College, Mangalore.
The first day’s session was taken up by Siddhartha Rao Kamalakara, a 4th year MIT student, who was one of the 2018 Google Summer of Code selects. He was also a previous Al team member of the self-driving car project and is a research enthusiast in the field of Machine Learning.
The participants were expected to bring in their laptops with the software of Python3, Tensorflow, Keras, Scikit Learn and Numpy pre-installed. The first day’s session, which lasted for two and a half hours, began with topics like the motivation behind machine learning, and why deep learning is required. Participants learned that Logistic Regression is a classification algorithm which is applied where the response variable is categorical and is used to find a relationship between features and the probability of a particular outcome.
The session then proceeded with the discussion of Neural Networks, which are powerful models that have a wide range of applications in the fields of medicine, security, banking, government, agriculture, and defence. They basically use the processing of the brain to develop algorithms that can be used to model complex patterns and prediction problems.
“The thing I found the best is that it was such a wonderful and rare opportunity where we were taught by a 4th year student,” said Priyam Agrawal, an organiser of the event. Later, the topics of the Cost function, Regularisation, and Gradient Descent were touched upon and explained perfectly with real-life examples and situations. That’s what grabbed the attention of the enthusiastic techies, who showed immense enthusiasm and active participation throughout the session.
The second day of IEEE’s Machine Learning workshop started early on Sunday, 5th August, with two sessions spanning across the entire day. The first session began at 10:30 am, taught by Ms Veena Maiyya, a research scholar at NITK, and faculty of I&CT at MIT, Manipal. The session picked up from the topics covered the previous day—mainly the principles of regression, one of the very basic and effective algorithms involved in machine learning. The lecture explained perceptrons and multi-layer perceptrons, leading into convolutional neural networks, and finally coming to an end by 12 pm.
The second session of the day resumed after lunch, taught by Siddhartha Rao Kamalakara, a fourth year student of MIT with considerable contribution to the Machine Learning research field. The afternoon session focused primarily on the working of neural networks, delving into their advantages and disadvantages and introductions to convolutional neural networks.
The workshop finally concluded with a discussion on the newer research areas in the field of ML—urging participants to take them up in the upcoming endeavors—as machine learning is widely understood as the future of technological advancement in countless fields today. The workshop provided an informal yet inquisitive temperament, being compelling as well as interactive. Being satisfied with the results, the organisers asserted hope for more such events to be conducted in the future, to further strengthen the growing research base in MIT Manipal.
Image Credits: The Photography Club, Manipal