It’s July 9th and I’m in the middle of my six week internship with the Air Force Research Laboratory and I decided to heed the advice of my mentor: You should create a blog. And now here we are, it’s my first post and I got a lot on my plate. As I’m writing this, I’m watching my not-very-well-trained A.I. agent hit random blocks in Minecraft. Actually, I’m able to interact with my agent and it’s pretty cool.
Check it out
This summer, I am a returning Wright Scholar and if you’re from southwest Ohio you might know what that program is. If you’re not, Wright Scholars is a relatively competitive program to do paid research in a field of your choosing. I, passionate about computer science and A.I., decided for those to be my fields. I was lucky enough to be tasked with two mentors, again one of whom told me I should start a blog, who are experts in the field. And I mean experts. They know so much about deep learning and reinforcement learning and all differnt subclasses of machine learning it’s crazy. As their intern and aprentice, I’ve been trying to soak as much information as possible. Last year, I did a supervised learning reserach project using GradCAM, an explainable A.I. technique used to highlight areas where a neural network is looking at, to help improve image detection of military vehicles. This year, I’m doing a reinforcement learning project based in Minecraft, as you got a sneak peek above, using mineRL, chainerRL, and OpenAI gym to train an agent to play Minecraft. Anywho, that’s about it for my first post, a bit of stream-of-conciousness and an intro to what this blog is sort of going to be about. We’ll see you in the next post!