STUDY GROUP ENROLMENT IS NOW CLOSED.
(UPDATED START TIME) Every Wednesday, starting on May the 2nd, from lunchtime from 12:00 to 2:00 pm for 8 weeks in central Sydney, SML will be working through Andrew NG’s recently released Deeplearning.ai course available on Coursera.
If you would like to get involved with a group of other like minded individuals to discuss or work through the homework exercises, we will be using the time to go over the homeworks together as a group.
If you would like to express interest in joining the group & find out the study groups location, please join our Sydney wide slack channel in order to find out more BY CLICKING THE LINK BELOW:
Attendees of the study group can expect to learn:
The basics of neural networks including forwards & backward propagation
Difference between train/ dev/ test sets
Bias & variance issues
Dropout, L2 regularization, vanishing/exploding gradients
Neural network Optimization techniques including: Mini batch gradient descent, stochastic gradient descent, Momentum, RMSprop, ADAM, learning rate decay etc
How to decide the train/dev/test split of your dataset
How to handle mis-matched training and dev/test data sets
How to apply transfer learning, multi-task learning and end-to-end learning.
Convolutional Neural Networks (CNNs): Learned about the terminologies used in CNN such as padding, stride and filter etc, basic operations of CNN such as pooling, and how to build multi-class classification using CNNs.
CNNs for object detection, CNNs for facial recognition & neural style transfer
Recurrent Neural Networks: Natural Language processing, word embeddings, LSTMs, Sequence models and attention mechanism
Feel free to bring along any questions you have about the lecture material or homeworks and we will work together to answer it. As teams, we will be working through the homeworks on a weekly basis.
Attendees will be expected to self study for 5 hours each week before attending the group - this study group is not for the faint of heart.
We have a team of professionals willing to assist throughout.
Teams that make it to the end will be automatically entered into our up & coming unannounced SML hackathon, with the opportunity to win prizes. More information will be sent via Slack to those who are successful in their application to the study group,
The team @ SML