Plans for the week of February 26-March 1

Dear all, we hope you've gotten time to enjoy the last weekend and look forward to a new week wit machine learning.

The plans this week are 

1. Finalizing discussion of Convolutional Neural Networks (CNNs), with an emphasis on how write your own code. Here, Eric Reber, who developed the code discussed in the lectures from last week, will discuss how he and Greg Kajda chose to develop the code. Eric is presently doing his master thesis at EPFL in Lausanne.

2. Thereafter we start with our new topic, which is about recurrent neural networks (RNNs)

3. Reading recommendations:

      a. Goodfellow, Bengio and Courville's chapter 10 from Deep Learning, https://www.deeplearningbook.org/

     b. Sebastian Rashcka et al, chapter 15, Machine learning with Sickit-Learn and PyTorch, https://sebastianraschka.com/blog/2022/ml-pytorch-book.html

     c. David Foster, Generative Deep Learning with TensorFlow, see chapter 5, see https://www.oreilly.com/library/view/generative-deep-learning/9781098134174/ch05.html

The last two books have codes for RNNs in PyTorch and TensorFlow/Keras.

The lecture notes for this week (some additions will be made during Monday) are at https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/pub/week7/ipynb/week7.ipynb

We remind you also that we have our regular lab session on Thursdays at 215pm-4pm, room FØ397.

Concerning the reports and group work, here are some simple guidelines:

1) Optimal groups are normally 2-3 participants,, but we have had larger groups as well, with up to five participants. As long as you can collaborate well, this is no problem. Larger groups tend to become more problematic to handle.  You can hand in an individual report based on your collaboration with the other group members (remember to state whom you collaborated with) or you can hand in a collective group report. In the latter case, everybody gets obviously the same score. 

2)  Length of reports: although the official website of the university of Oslo states that the maximum length should not exceed ten pages, we are pretty relaxed here. If you wish to write more, feel free to do so.  You will not be penalized if you write a longer report.  This is very much up to you. But it is important to strike a good balance between length and the amount of work that corresponds to 10 ECTS!!

You can find a detailed guideline on how to write a scientific report (and how we give a score and feedback on the projects) at  https://github.com/CompPhysics/MachineLearning/blob/master/doc/Projects/EvaluationGrading/EvaluationForm.md

Best wishes to you all,

Keran, Morten and Ruben