MIST101-3: Artificial Neural Networks

2020/2021: 1 | 2 | 3 | 4 | 5


Image of MIST101-3

We are sure you must have heard about artificial neural networks, a paradigm of machine learning that has recently gained significant popularity and rightly so. But did you know that our brain in itself is a complicated biological neural network and the artificial neural network is inspired by it?

In this workshop, we will develop this idea further and explain the mathematical formulation and working of an artificial neural network along with the similarities it shares with our human brain. We will dive deeper into the concepts of activation functions, gradient descent, and backpropagation to understand the working of an artificial neural network in detail.

We strongly believe that the best way to learn a concept is to apply it and hence, we have organized two competitions where you can apply the concepts you have learned and challenge yourself to master these concepts! The first competition will help you understand hyperparameter tuning and how it affects model performance. The second competition challenges you to alter the model complexity and in the process teach the relationship between model complexity & model accuracy. Further details about the competitions will be provided during the workshop and the submission details can be found below.

Did I hear gift cards? Yes, of course, once again, we are partnering with Amazon Prime Student to provide gift cards worth 25 CAD each when you answer them correctly* ! So stay tuned for our upcoming workshop which will be released on October 15th, 2020.

Wait a minute, what time is it?…. We value the fact that you are busy, and hence we’ll upload the recorded videos on our YouTube channel, so that you can watch it when you have time!

Presentation slides are here and the Collab notebook can be found here.

Event Details

Topics to be covered:

  1. What is a neural network?
  2. BNN (Biological Neural Network)
  3. ANN (Artificial Neural Network)
  4. Activation functions
  5. Feedforward neural network
  6. Backpropagation

How to join our contests:

  • Contest - I: Hyperparameter tuning

    • Take the screenshot of the accuracy you achieved (Full-screen snapshot).
    • Write a short paragraph explaining the hyper-parameters you changed to achieve it.
    • Submit the screenshot and your answer, via email to roland[dot]gao[at]mail[dot]utoronto[dot]ca
  • Contest - II: 2-Layer ANN

    • Implement the 2-layer of ANN using Google Colab
    • Submit the Google Colab link of your code and the accuracy you achieved, via email to roland[dot]gao[at]mail[dot]utoronto[dot]ca.

We’ll contact you if you win!

References:

Nagyfi, R. (2018, September 04). The differences between Artificial and Biological Neural Networks. Retrieved October 08, 2020, from https://towardsdatascience.com/the-differences-between-artificial-and-biological-neural-networks-a8b46db828b7

Back Propagation Neural Network: Explained With Simple Example. (n.d.). Retrieved October 08, 2020, from https://www.guru99.com/backpropogation-neural-network.html

7 Types of Activation Functions in Neural Networks: How to Choose? (n.d.). Retrieved October 08, 2020, from https://missinglink.ai/guides/neural-network-concepts/7-types-neural-network-activation-functions-right/

Artificial Neural Network - Building Blocks. (n.d.). Retrieved October 08, 2020, from https://www.tutorialspoint.com/artificial_neural_network/artificial_neural_network_building_blocks.htm

* An Amazon gift card worth 25$ will be provided to the top submissions in each of the two competitions.