Skip to content

JA6. Neural Networks

Statement

Your learning journal entry must be a reflective statement that considers the following questions:

1. Describe what you did

This was the 6th week of this course, it was about the basics of neural networks. I started the week as usual by trying the self-quiz then I watched the lecture videos and started to read the text; the text size for this week was reasonable and I was able to skim it. Later, I did the discussion assignments, graded quiz, and finally I am preparing the learning journal.

Unfortunately, I had personal circumstances this week that prevented me from doing the programming assignment, but I will try to do it later.

2. Describe your reactions to what you did

I have a background in pharmacy, which involves a fair amount of biology. The workings of the nervous system have always fascinated me, so delving into neural networks (a computing system inspired by biological neurons) was satisfying. I liked how artificial neurons mimic biological neurons in processing information and passing nerve impulses (Gershenson, 2003).

3. Describe any feedback you received or any specific interactions you had. Discuss how they were helpful

I have not received any feedback worth mentioning, possibly due to my late participation in the discussion forums. However, I have been able to learn from the posts of other students.

4. Describe your feelings and attitudes

I felt good about the topic at small scale; but I don’t think I am ready to understand the topic at large scale. Also, the simulator did not work on my Mac device which was one of the reasons I missed the programming assignment. I felt frustrated as tried to find an alternative simulator but I could not find any (within the time constraints that I had).

5. Describe what you learned

The required readings mentioned biological neural networks, artificial neural networks and their applications, feed-foreword networks, feed-back networks, self-organizations (Peterson & Rögnvaldsson, 1991).

We also learned about hidden layers, bias nodes, logic gates, cost function, and gradient back-propagation (Cowan, 2013).

We also learned about similarities and differences between artificial neural networks and biological neural networks, the components of neural networks, and how weightings are adjusted to improve the performance of the network (Gershenson, 2003).

6. What surprised me or caused me to wonder?

I liked how neural networks seems complicated when looking at at a diagram or doing the math for a large number of nodes; however, reducing the scale of the problem at first made diagrams and calculations much easier to understand. I also liked how neural networks can be used to solve problems that are not easily solved by other methods; and its ability to adjust weightings to improve its performance (Cowan, 2013).

7. What happened that felt particularly challenging? Why was it challenging to me?

The text reading from (Peterson & Rögnvaldsson, 1991) was challenging as it is a bit old and lengthy with very little diagrams and colors.

8. What skills and knowledge do I recognize that I am gaining?

I have gained theoretical knowledge about neural networks, how they mimic biological neural networks, and how they use back-propagation to adjust weightings to improve their results.

9. What am I realizing about myself as a learner?

I realized that it is hard to me to understand theoretical topics without practical examples; unfortunately, the simulator did not work for me and it was hard to find an alternative simulator. I also realized that real-word problems are much more complex than the examples that we are working with in this course.

10. In what ways am I able to apply the ideas and concepts gained to my own experience?

Given my current role as a software engineer, neural networks provide another tool for solving small classification and regression problems in my projects; but most importantly, it gives me the necessary terminology to understand and communicate with data scientists and machine learning engineers in the company.

11. Describe one important thing that you are thinking about in relation to the activity

I had created a recommendation algorithm in my work that uses the idea of inputs, weightings, hidden layers, and outputs; however, the weightings were hard coded in the code and were not adjusted automatically. I am thinking of how to make those weightings adjustable to improve the usefulness of the algorithm.

References