Outline¶
The overview already specified it, but just to be sure: the workshop will be split into two distinct parts. While the first half (and few hours) will consist of going through the theoretical background of “classic” machine learning and deep learning, the second half will entail a dedicated hands-on session within which folks will evaluate the discussed topics and their feasibility for their own data. As mentioned before, these resources will only contain the materials for the first part, as the second will include work on data brought by the workshop attendees. In general, we will aim for 2 h per block, that is 2 h for “classic” machine learning and 2 h for deep learning. Within each we will go from basic concepts and building blocks to model evaluations and important problems. Our very optimistic schedule looks as follows (all times in EST):
Welcome (9 AM - 9:30 AM)¶
Time slot |
Topic |
|---|---|
9 AM - 9:15 AM |
General hello, introduction round and organization (9 AM - 9:15 AM) |
9:15 AM - 9:30 AM |
Models, AI and all other buzz words (9:15 AM - 9:30 AM) |
The content I - theoretical background (9:30 AM - 12 PM)¶
Time slot |
Topic |
|---|---|
9:30 AM - 10 AM |
“Classic” machine learning - supervised or unsupervised, model types |
10 AM - 10:15 AM |
yoga/dance break |
10:15 AM - 10:45 AM |
“Classic” machine learning - model evaluation & cross-validation |
10:45 AM - 11:15 AM |
“Classic” machine learning - model tuning & biases |
11:15 AM - 11:30 AM |
yoga/dance break |
11:30 AM - 12 PM |
Deep learning - basics & architectures |
Lunch (12 PM - 1 PM)¶
The content I - theoretical background (1 PM - 2:15 PM)¶
Time slot |
Topic |
|---|---|
1 PM - 1:30 PM |
Deep learning - how to build and train your neural network |
1:30 PM - 2 PM |
Deep learning - model tuning & biases |
2 PM - 2:15 PM |
yoga/dance break |
The content II - hands-on (2:15 PM - 4 PM)¶
Time slot |
Topic |
|---|---|
2:15 PM - 2:30 PM |
Dataset blitz |
2:30 PM - 3:45 PM |
Free hacking |
3:45 PM - 4 PM |
Lessons learned, Q&A |