Annotated Course Schedule and Outline

Note: papers that report on models we will be implementing in the assignments are denoted by asterisks.

Week 1: Introduction, Hopfield Networks

• Discussions:

  • What is memory?

  • What does it mean to build a “model” of memory?

  • Are neural networks like biological brains?

• Hebbian learning and Hopfield networks

  • Readings:

    • Hopfield (1982)

    • Hopfield (1984)

    • Amit et al. (1985)*

• Discussion: Hopfield network simulations (storage capacity, cued recall, contextual drift)

• Assignment 1: Explore Hopfield Networks

Weeks 2–3: Free recall, Short Term and Long Term Memory

• Discussions:

  • free recall and memory search

  • naturalistic memory tasks

• Readings:

  • Atkinson and Shiffrin (1968)*

  • Chen et al. (2016)

  • Heusser et al. (2021)

• Data science primer:

  • Where to find behavioral datasets: Penn Behavioral Data Archive, OpenCogData, OpenNeuro, UCLA Psychological Dataset Archive, Context Lab

  • Web scraping with requests and Beautiful Soup

  • Data manipulation with Pandas

  • Text analyses with Scikit-learn, NLTK, and HuggingFace Transformers

• Assignment 2: Build the Search of Associative Memory Model

Weeks 4–5: Temporal Context and Multi-Timescale Models

• Discussion: the temporal scales of memory, event boundaries, and situation models

• Readings:

  • Howard and Kahana (2002)

  • Polyn et al. (2009)*

  • Baldassano et al. (2017)

  • Honey et al. (2012)

  • Manning et al. (2014)

  • Ranganath and Ritchey (2012)

  • DuBrow and Davachi (2016)

  • Zacks and Tversky (2001)

  • Zwann and Radvansky (1998)

  • Brunec et al. (2018)

• Assignemnt 3: Build the Context Maintenance and Retrieval Model

Week 6–7: Laplace Transforms

• Discussion: is TCM really multi-timescale?

• Discussion: Introduction to the Laplace Transform (and its inverse) and its relevance to memory

• Readings:

  • Shankar and Howard (2012)*

  • Manning (2024)

• Assignment 4: Implement the Laplace Temporal Context Model

Week 8: Biologically Inspired Network Models

• Discussion: what does “biologically inspired” mean in practice?

• Readings:

  • McClelland et al. (1995)

  • Rumelhart et al. (1986)

  • O’Reilly and Norman (2002)

  • Schapiro et al. (2017)

Week 9: Recurrent networks, LSTM networks, Transformers

• Readings:

  • Schuster and Paliwal (1997)

  • Hochreiter and Schmidhuber (1997)

  • Radford et al. (2019)

    • Tutorial video: Let’s build GPT: from scratch, in code, spelled out

• Assignment 5: Final Project

Week 10: Final project presentations

• Discussion: ad-hoc discussions and demos of final projects

• Final projects are due on the last day of class at 11:59PM Eastern Time