Skip to content

Latest commit

 

History

History
53 lines (44 loc) · 5.18 KB

Module 4: Model Tuning & Recommenders.md

File metadata and controls

53 lines (44 loc) · 5.18 KB

Module 4: Model Tuning & Recommenders

Class 8: Model Tuning
Each team gives a 5-minute presention on an assigned section of Deep Learning Tuning Playbook or ISLP Chapter 10
Section 1 Presentation Slidedeck
Section 2 Presentation Slidedeck

Midterm Review
Python Functions
Colab: Python Functions
Class Inheritance
Colab: Class Inheritance

Class 9: Mid-term Exam & Guest Speaker
8:30-10:20am: Mid-term Exam in Brinkley (both sections)
10:30-11:20am Guest Speaker Zhen Zeng on Zoom in Brinkley (both sections)
11:30am Pizza! Class ends for both sections

Guest Speaker Research Presentation for BUAD5742: Artificial Intelligence (MSBA)
Title: Visual Perspectives on Financial Time Series
Location: on Zoom in Brinkley, Miller Hall, Mason School of Business, William & Mary
Speaker: Dr. Zhen Zeng
Dr. Zeng is an AI Research Lead on the AI Research team at J.P. Morgan. Her research interests are novel representations of financial data, computer vision, robotics and LLMs. Her work at J.P. Morgan features time series analysis and prediction with visual representations, and enabling LLM agents for real-world tasks. Before joining J.P. Morgan, she worked in robotics and enabling agents to perceive, understand and make decisions in the physical world through probabilistic modeling and machine learning. She received a Ph.D. in Electrical and Computer Engineering from the University of Michigan. She has won best paper awards in Cognitive Robotics at IEEE International Conference on Robotics and Automation (ICRA) 2020, and Mobile Manipulation in International Conference on Intelligent Robots and Systems (IROS) 2021.

Complete these readings/videos before next class meeting
Recommendation Systems, Google's Advanced Course on Machine Learning
Collaborative filtering
Recommenders Tutorial by Tensorflow

Optional References
Recommended For You: How machine learning helps you choose what to consume next by Jenifer Wei Case Study: YouTube's recommendation system
Case Study: The history of Amazon's recommendation algorithm
Case Study: How Netflix's Recommendation System Works
Colab: Item-to-Item collaborative filtering

Optional Fun Stuff
How YouTube Knows What You Should Watch
Let's Make a Movie Recommendation System

Class 10: Recommenders
M4-1 Recommenders
M4-2 Vector Similarity
Colab: Vector Similarity
M4-3 Candidate Generation
Colab: TRFS Recommenders

Complete these readings/videos before next class meeting
Reinfrocement Learning by MLU
TF Reinforcement Learning
TF Introduction to Multi-Armed Bandits
Multi-Armed Bandits and the Stitch Fix Experimentation Platform