Hi, I’m Mayeesha. I’m from Bangladesh.
How did you find out about the course? Have you used your knowledge from part 1 on any fun projects yet?
– I found out about the course when I was searching for materials on transfer learning in Youtube. I was doing the capstone project for the Udacity Machine Learning Nanodegree and I had chosen Kaggle’s fishery competitions for that, mostly because the data set was small. Even if I don’t have any particular passion for fishes, the project started getting interesting after I accumulated some knowledge in deep learning and started understanding the overall conversations around DL models. It seemed like even if there’s many good theoretical courses including CS231 for CNN’s, practical courses are super rare, so after I found the youtube lectures I started watching them.
What are you hoping to do with deep learning?
– So far I’ve felt like the entire conversation about deep learning was basically about one topic when I was doing courses at udacity, which is self driving cars. Right now I don’t even know driving, and I am not particularly interested in the self driving cars, except the part that I’d like to ride them if possible to see how it goes. After watching the video on teaching philosophy I was sold on the course.
I wanted to do well in the kaggle competitions that use DL, but I felt those were out of my league. So that was one of my motivations. Right now I’m ranking around 45% in the Fishery one, (I’ve not tried the model with the fully convolutional layers yet though as shown in the lecture), so I feel more confident about my skills in implementing neural networks. My experience in this area is still limited and I’m still trying to understand things, but I want to use deep learning on animation, and basically focus more on story-telling.
Now, there’s not much info available (and I’ve probably not researched well enough yet), how to use Machine learning in the field of animation and I’m still just trying to explore it. To be really honest, I want to know how to make more interesting, entertaining products using ML and specially DL. My general style is basically learning techniques first and thinking about projects later with them. Even in this network visualization project I did before, I learnt about networks and Gephi first and later used it when the opportunity came. It’s a RPG like thing for me, if I know the techniques, I can use them when needed to ‘unlock’ cooler projects. Also I feel like this whole area of ‘using deep learning in creative, artistic project’ area is super super new for me, it’d be a good idea to just explore the field first for a while.
What are you passionate about?
– Reading books, writing, food, cats(I’ve 6 cats), watching anime’s/movies, general social psychology, economics, lots of stuff. I like social networks, human interactions in a broad scale, pretty much random things.
What’s something that not many people know about you?
— I’m not that much anti-social, regardless of my general vibe offline. I’m just introverted. I really like animals and inanimate objects more than human beings. I think I also have a habit of anthropomorphizing things which kind of explains my interest in animation, animals and even robot ethics. I dislike hyper structured environments.