WAT.ai Events
\
Dive into Deep Learning Workshop (part 2)
Wednesday, October 4th, 5-7pm. In E5 2004
Prerequisites:
Same as Part 1, but concepts covered in Neural Networks for Novices are assumed to be prerequisite knowledge.• Explore more complex model architectures, block structures, and applications of neural networks that are common in industry and academia
• Investigate how deep learning can be applied to different data modalities and for a wider variety of machine learning tasks
• Apply this knowledge to implement a deep learning model in PyTorch, based on technical specifications
Previous Events
General Kickoff Meeting
Wednesday September 13th, 5-7pm. In E5 2004
• Meet the team• Learn more about our educational workshops
• Hear more regarding our new project
• Answer questions regarding recruitment process
A.I. Literacy Workshop
Thursday, September 14th, 5-6pm. In E5 2004
Prerequisites:
None• Provide examples of A.I. applications in industry and identify the appropriate tools for solving various kinds of problems
• Introduce the challenges of working with A.I. systems and the importance to data and model literacy in the machine learning engineering process
Data Preprocessing Workshop
Wednesday, September 20th, 5-7pm. In E5 2004
Prerequisites:
Some Python or equivalent experience, including a basing understanding of python syntax, loops, conditional statements, functions, and data types.• Identify outliers, handle missing values, and perform other common data operations such as normalization, interpolation, and filtering
• Understand the intuition behind various preprocessing techniques for both categorical and continuous features
• Apply EDA and data preprocessing techniques to a novel data set without context
Classical ML Workshop
Thursday, September 21st, 5-7pm. In E5 2004
Prerequisites:
Some Python or equivalent experience, including a basing understanding of python syntax, loops, conditional statements, functions, and data types. Some background in numerical computing (MATLAB, R, NumPy or similar) and familiarity with linear algebra would be helpful.• Provide intuition for various classical machine learning techniques regarding their complexity, performance, and effectiveness in the context of different applications
• Explore concepts such as feature selection, model selection, hyperparameter tuning, performance metrics, and the bias/variance trade-off
• Apply this knowledge to a real-world dataset in a competition-style activity
Neural Networks for Novices Workshop (part 1)
Sunday, October 1st, 1-3pm. In E5 2004
Prerequisites:
• Python experience, including an understanding of python syntax, loops, conditional statements, functions, and data types in python• Some background in numerical computing - MATLAB, R, NumPy, or similar, and an understanding of vectors, matrices, and relevant linear algebra concepts
• An understanding of model selection, train/test split, performance metrics and other concepts covered in the session on Classical Machine Learning
• Define and discuss hyperparameters in the context of deep learning models, including learning rate, batch size, epochs, layers, hidden units, optimizers, and activation functions
• Interpret loss and accuracy curves to identify overfitting during the training process
• Apply this knowledge to a real-world dataset using TensorFlow