My AI Saturdays (Lagos) Journey -Week 7

It’s week 7 and the journey continues. I woke up with two things on my mind — AI Saturdays meetup and the Google Local Guides meetup. Had to choose AI Saturdays for obvious reasons and there will always be another Local Guides meetup.

We had a review of last week’s session and jumped right into the topic of the day which was Back Propagation with Neural Networks. George Igwebe and Lawrence Francis took us through the math and codelab sessions respectively.

George explaining computational graph and backpropagation

According to Wikipedia:

Backpropagation is a special case of an older and more general technique called automatic differentiation. In the context of learning, backpropagation is commonly used by the gradient descent optimization algorithm to adjust the weight of neurons by calculating the gradient of the loss function. This technique is also sometimes called backward propagation of errors, because the error is calculated at the output and distributed back through the network layers.

Lawrence getting set for a live code session on backpropagation.

After the sessions, we were grouped into 7 teams and the goal is to use frameworks like Tensorflow to solve a multiclass problem, working with the MNIST dataset or Fashion MNIST dataset.

My team members are :

1. Oladipo Anjolaiya (Lead)
2. Ifeoluwa Jesuniyi
3. Jennifer Akwari
4. Joshua Akande
5. Timi Balogun
6. Ibrahim Gana(me)

As beginners it is going to be tasking but we are ready for the challenge. Honestly, for me it is not about winning, but about learning(Nevertheless, we are still bringing our A-game).

Remote session with Prof. Thomas G. Dietterich.

The guest for this weekend’s intermediate class was Prof. Thomas G. Diettrich, Professor of Computer Science at Oregon State University.

According to Wikipedia:

He is one to the founders of the fields of machine learning. He served as Executive Editor of Machine Learning (journal) (1992–98) and helped co-found the Journal of Machine Learning Research. In response to the media’s attention on the dangers of artificial intelligence, Dietterich has been quoted for an academic perspective to a broad range of media outlets including National Public Radio, Business Insider, Microsoft Research, CNET, and The Wall Street Journal.[1]

During his introduction to the class, I froze when he said he has been working on machine learning as far back as 1977. He spoke about Robust Machine Learning with confidence and I must say it was an insightful session.

Click to watch the session: Remote session with Prof. Thomas G. Dietterich