CS 370 - Spring 2025. 4/2/2025

[Home]

Welcome to CS 370!

Video of the Day

Quantitative Fixation See related article: Does counting change what counts? Quantification fixation biases decision-making, Linda W. Chang, Erika L. Kirgios, Sendhil Mullainathan and Katherine L. Milkman.

Also, check out the podcast: Choiceology, which has a strong behavioral finance slant.

Also, Monday I discussed qualitative arithmetic and the idea that agents prefer numbers to be either high or low. There is another possibility. Consider the following: weight, height, temperature, blood pressure, pulse, cholesterol.

I hereby solicit suggestions for the video of the day. Please email me your ideas with explanations. Selected entries will win 5 homework points. If your video is played at the beginning of class, you must also briefly explain something about the video and something about yourself - in person.

Canvas Quiz of the Day (need daily password)

Most days, there will be a simple canvas quiz related to the lecture. You need a password to activate the quiz, which I will provide in class. These quizzes will count toward your class participation grade. The quiz is available only during class. You get full credit for class participation by completing half of the quizzes.

Click for today's quiz.

Lecture 19: Learning

Announcements

  • Information Society Project Yale Law School. Weekly Events

  • Yale Pitchbook Mixer, Thursday April 3rd, 8:30pm - 10:30pm, HQ Room 131 (Student Lounge)
    Yale Undergraduate Capital Partners (YUCP) is officially launching YalePitchbook.com - the first platform connecting Yale startups with investors, talent, and resources, and building a robust ecosystem of founders, investors, operators, and startup service providers in New Haven & beyond! The platform is backed by Yale Ventures, Yale School of Management, and Tsai Center for Innovative Thinking at Yale (Tsai CITY).

    To celebrate the launch, YUCP and YES are hosting a launch party and entrepreneurship mixer, sponsored by Xfund, from 8:30-10:30 pm on Thursday, April 3, at the HQ Lounge, with free drinks and food!

    Expect founders, VCs, angels, joiners, current students, alumni, and friends of the Yale startup ecosystem all in one room - plus a chance to explore the startups shaping the future from within Yale’s walls. Whether you're looking to invest, collaborate, or just get inspired, this is the event to be at.

    RSVP now: https://lu.ma/5d47dx6x. Flyer attached!

    Administrivia

  • I have office hours Wednesdays from 4-6 p.m. via Zoom, meeting ID 459 434 2854.

  • The TF's office hours are posted on Ed Discussion.

  • I am available for lunch on Mondays at 1 pm in Morse.

  • Homework assignments: [Assignments]. hw6 is now available. See Ed Discussions for links to other Titanic video tutorials. Also, see Tyler's post about using FastX on the zoo for running jupyter notebooks remotely.

    I have reviewed all the project proposals. Let me know if you did not get an OK.

    Asides from previous lectures

    AI in the news

  • Tinder’s Flirting AI Chatbots Play Hard to Get, Bloomberg, April 2, 2025.

  • Access to the Atlantic
  • Access to Economist (Economist.com)
  • Access to Financial Times
  • Access to Wall Street Journal from Yale.
  • Q and AI Bloomberg.
  • Access to Bloomberg.com from Yale.

    Lecture: Learning

    Irises, Vincent Van Gogh.

  • Readings: chapters 19-21
  • 2019 Scassellati Slides:
  • AIMA Slides:

  • 0324.ipynb 0324.html Perceptron ***
  • Ensemble learning
  • Boosting

  • Machine learning with SciKit Learn *** 
    pip install scikit-learn
  • compare various sklearn algorithms jupyter notebook HTML file. ***
  • sklcompare.py Used in previous. Error in numpy versions. Sigh.

  • learning_apps.ipynb learning_apps.html See MNIST data (Modified National Institute of Standards and Technology.) Combines census and high school student samples. ***

    Neural Networks

    Perceptrons Minsky and Papert, 1969.

  • Boolean Functions with Perceptrons

  • Neural Networks and Deep Learning, Michael Nielsen. ***

  • MNIST Neural Network from Nielsen. jupyter notebook ***
  • Neural Network playground ***

  • Like Boolean circuits (from CS 201). A feed-forward neural network is like a combinational circuit or a directed acyclic graph, with designated input and output nodes. However, instead of Boolean values of 0 or 1, there is a continuous value between 0 and 1. The usual activation functions include logistic (sigmoid), tanh, ReLU (rectified linear unit), or linear. (See table). All are monotonically nondecreasing, which means that their derivatives are nonnegative.
  • A recurrent neural network is like a sequential circuit, which includes feedback. It is a dynamic system with internal state or memory.
  • A Transformer architecture is derived from the 2017 paper Attention is all you need from Google researchers, which lead to the LLM development.
  • neural_nets.ipynb neural_nets.html

    Future topics: NLP and LLMs

  • Moore's Law vs the More Law.
  • Hands-on large language models : language understanding and generation Jay Alammar, O'Reilly, 2024. (Yale library online book).
  • NLP Progress Repository to track the progress in Natural Language Processing (NLP), including the datasets and the current state-of-the-art for the most common NLP tasks.
  • C4 (Colossal Clean Crawled Corpus)
  • Efficient Estimation of Word Representations in Vector Space the word2vec paper, by Jeff Dean and the Google guys. 2011.
    [Home]