How do you start an independent research?

How to Start Doing Independent Research In Your Undergrad

1. Talk to professors. Tell them about what you are passionate about and let them know ahead of time if there is something you’d like to pursue as a specific career development.
2. Decide on a topic.
3. Ask the professor to allow you to do a not-for-credit project.

How do I start research on machine learning?

Start with a clear idea of why you want to research a given machine learning algorithm, and then pick those sources that can best answer the questions that you have. There are 5 different sources that you can use in your research of a machine learning algorithm, we will review each in turn.

How often are new papers published in deep learning?

With hundreds of papers being published every month, anybody who is serious about learning in this field cannot rely merely on tutorial-style articles or courses where someone else breaks down the latest research for him/her. How do you continue the learning after you have consumed that book or completed that amazing online course on Deep Learning?

How can I prepare for a career in deep learning?

But make sure you learn to program. Here are some other things you can do to help prepare for a career in deep learning: Take an AI-related problem you are passionate about and think about it on your own. Once you have formed your own idea of it, start reading the literature on the problem.

Which paper rekindled all the interest in deep learning?

This is the paper that rekindled all the interest in Deep Learning. Authored by Alex Krizhevsky, Ilya Sutskever, and Geoffrey Hinton, and titled ImageNet Classification with Deep Convolutional Networks, this paper is regarded as one of the most influential papers in the field.

Is there a difference between theory and implementation papers in deep learning?

At least within deep learning, there are papers more focused on theory and others more focused on implementations. In terms of how you might want to approach your own interests and work, it may be easier to start off with implementations, as the effects of these papers are more concrete.