Machine Learning for Structured Data

Assignments

There will be 6 homework assignments during the semester. The assignments will consist of both theoretical and programming problems. Homework assignments will be released via a Piazza announcement explaining where to find the handout, starter code, LaTeX template, etc.

• Homework 1: PyTorch Primer: MLP for Sequence Tagging [natural language processing] (written / programming)
  [Handout]
• Homework 2: Learning to Search for RNNs [speech recognition] (written / programming)
  [Handout]
• Homework 3: General-graph CRF Module for PyTorch [natural language processing] (written / programming)
  [Handout]
• Homework 4: Gibbs Sampling [topic modeling] (written / programming)
  [Handout]
• Homework 5: Variational Inference (written / programming)
  [Handout]
• Homework 6: Advanced Topics (written)
  

Tentative release dates and due dates are listed on the Schedule page.

Exams

There will be two exams. The links to the Practice Problems and Exam Exit Polls will be provided below.

• Practice Problems – Midterm Exam
  [Practice] [Solutions]

• Practice Problems – Final Exam
  [Practice] [Solutions]

Mini-Project

The mini-project is required for 10-618 students only.

The course mini-project affords an opportunity to explore a learning / inference technique of your choosing. The application and dataset will be provided (in the style of a Kaggle competition) and you will implement a structured prediction approach with the goal of achieving both efficiency and accuracy.

The mini-project must be completed by a team of two 10-618 students.

Details about the project milestones and deliverables can be found here:
[Handout]