The Human Element in Machine Learning w Catherine D’Ignazio, Jacob Andreas & Harini Suresh (S3:E5)

MIT OpenCourseWare · Beginner ·📐 ML Fundamentals ·4y ago
When computer science was in its infancy, programmers quickly realized that though computers are astonishingly powerful tools, the results they achieve are only as good as the data you feed into them. (This principle was quickly formalized as GIGO: “Garbage In, Garbage Out.”) What was true in the era of the UNIVAC has proved still to be true in the era of machine learning: among other well-publicized AI fiascos, chatbots that have interacted with bigots have learned to spew racist invective, while facial-recognition software trained solely on images of white people sometimes fails to recognize people of color as human. In this episode, we meet Prof. Catherine D’Ignazio of MIT’s Department of Urban Studies and Planning (DUSP) and Prof. Jacob Andreas and Harini Suresh of the Department of Electrical Engineering and Computer Science. In 2021, D’Ignazio, Andreas, and Suresh collaborated as part of the Social and Ethical Responsibilities of Computing initiative from the Schwartzman College of Computing in a project to teach computer science students in 6.864 Natural Language Processing to recognize how deep learning systems can replicate and magnify the biases inherent in the data sets that are used to train them. Relevant Resources: MIT OpenCourseWare https://ocw.mit.edu/index.htm?utm_source=youtube&utm_medium=shownotes&utm_campaign=chalkradio&utm_term=s3e5 The OCW Educator Portal https://ocw.mit.edu/educator?utm_source=youtube&utm_medium=shownotes&utm_campaign=chalkradio&utm_term=s3e5 Share your teaching insights https://forms.gle/XBwUwqn35abSdjNs8 Case Studies in Social and Ethical Responsibilities of Computing https://ocw.mit.edu/resources/res-tll-007-case-studies-in-social-and-ethical-responsibilities-of-computing-fall-2021/? utm_source=youtube&utm_medium=shownotes&utm_campaign=chalkradio&utm_term=s3e5 SERC website https://computing.mit.edu/cross-cutting/social-and-ethical-responsibilities-of-computing/ Professor D’Ignazio’s faculty page https://dusp.m
Watch on YouTube ↗ (saves to browser)
Sign in to unlock AI tutor explanation · ⚡30

Playlist

Uploads from MIT OpenCourseWare · MIT OpenCourseWare · 0 of 60

← Previous Next →
1 21. Post Trade Clearing, Settlement & Processing
21. Post Trade Clearing, Settlement & Processing
MIT OpenCourseWare
2 10. Financial System Challenges & Opportunities
10. Financial System Challenges & Opportunities
MIT OpenCourseWare
3 7. Technical Challenges
7. Technical Challenges
MIT OpenCourseWare
4 3. Blockchain Basics & Cryptography
3. Blockchain Basics & Cryptography
MIT OpenCourseWare
5 19. Primary Markets, ICOs & Venture Capital, Part 1
19. Primary Markets, ICOs & Venture Capital, Part 1
MIT OpenCourseWare
6 1. Introduction for 15.S12 Blockchain and Money, Fall 2018
1. Introduction for 15.S12 Blockchain and Money, Fall 2018
MIT OpenCourseWare
7 Chalk Radio, A Podcast about Inspired Teaching at MIT (Teaser)
Chalk Radio, A Podcast about Inspired Teaching at MIT (Teaser)
MIT OpenCourseWare
8 Nuclear Gets Personal with Prof. Michael Short (S1:E1)
Nuclear Gets Personal with Prof. Michael Short (S1:E1)
MIT OpenCourseWare
9 How Africa Has Been Made to Mean with Prof. Amah Edoh (S1:E2)
How Africa Has Been Made to Mean with Prof. Amah Edoh (S1:E2)
MIT OpenCourseWare
10 Making Deep Learning Human with Prof. Gilbert Strang (S1:E3)
Making Deep Learning Human with Prof. Gilbert Strang (S1:E3)
MIT OpenCourseWare
11 Social Impact at Scale, One Project at a Time with Dr. Anjali Sastry (S1:E4)
Social Impact at Scale, One Project at a Time with Dr. Anjali Sastry (S1:E4)
MIT OpenCourseWare
12 Film is for Everyone with Prof. David Thorburn (S1:E5)
Film is for Everyone with Prof. David Thorburn (S1:E5)
MIT OpenCourseWare
13 Lecture 12: Aircraft Performance
Lecture 12: Aircraft Performance
MIT OpenCourseWare
14 Lecture 3: Learning to Fly
Lecture 3: Learning to Fly
MIT OpenCourseWare
15 Lecture 13:  Interpreting Weather Data
Lecture 13: Interpreting Weather Data
MIT OpenCourseWare
16 Lecture 21: Weather Minimums and Final Tips
Lecture 21: Weather Minimums and Final Tips
MIT OpenCourseWare
17 Hand-on, Minds On with Dr. Christopher Terman (S1:E6)
Hand-on, Minds On with Dr. Christopher Terman (S1:E6)
MIT OpenCourseWare
18 Part 4: Eigenvalues and Eigenvectors
Part 4: Eigenvalues and Eigenvectors
MIT OpenCourseWare
19 Part 5: Singular Values and Singular Vectors
Part 5: Singular Values and Singular Vectors
MIT OpenCourseWare
20 Part 3: Orthogonal Vectors
Part 3: Orthogonal Vectors
MIT OpenCourseWare
21 Part 2: The Big Picture of Linear Algebra
Part 2: The Big Picture of Linear Algebra
MIT OpenCourseWare
22 Part 1: The Column Space of a Matrix
Part 1: The Column Space of a Matrix
MIT OpenCourseWare
23 Intro: A New Way to Start Linear Algebra
Intro: A New Way to Start Linear Algebra
MIT OpenCourseWare
24 9. Chromatin Remodeling and Splicing
9. Chromatin Remodeling and Splicing
MIT OpenCourseWare
25 28. Visualizing Life - Fluorescent Proteins
28. Visualizing Life - Fluorescent Proteins
MIT OpenCourseWare
26 20. Roth's theorem III: polynomial method and arithmetic regularity
20. Roth's theorem III: polynomial method and arithmetic regularity
MIT OpenCourseWare
27 8. Szemerédi's graph regularity lemma III: further applications
8. Szemerédi's graph regularity lemma III: further applications
MIT OpenCourseWare
28 19. Roth's theorem II: Fourier analytic proof in the integers
19. Roth's theorem II: Fourier analytic proof in the integers
MIT OpenCourseWare
29 12. Pseudorandom graphs II: second eigenvalue
12. Pseudorandom graphs II: second eigenvalue
MIT OpenCourseWare
30 1. A bridge between graph theory and additive combinatorics
1. A bridge between graph theory and additive combinatorics
MIT OpenCourseWare
31 Special Episode: Teaching Remotely During Covid-19 with Prof. Justin Reich
Special Episode: Teaching Remotely During Covid-19 with Prof. Justin Reich
MIT OpenCourseWare
32 Spring 2020 Update from Dean Rajagopal
Spring 2020 Update from Dean Rajagopal
MIT OpenCourseWare
33 S1E7: Unpacking Misconceptions about Language & Identities with Prof. Michel DeGraff
S1E7: Unpacking Misconceptions about Language & Identities with Prof. Michel DeGraff
MIT OpenCourseWare
34 Climate 101 Live
Climate 101 Live
MIT OpenCourseWare
35 Welcome for Volunteers (for EarthDNA's Climate 101)
Welcome for Volunteers (for EarthDNA's Climate 101)
MIT OpenCourseWare
36 Learning to Fly with Drs. Philip Greenspun & Tina Srivastava (S1:E8)
Learning to Fly with Drs. Philip Greenspun & Tina Srivastava (S1:E8)
MIT OpenCourseWare
37 Thinking Like an Economist with Prof. Jonathan Gruber (S1:E9)
Thinking Like an Economist with Prof. Jonathan Gruber (S1:E9)
MIT OpenCourseWare
38 2. Cyber Network Data Processing; AI Data Architecture
2. Cyber Network Data Processing; AI Data Architecture
MIT OpenCourseWare
39 1. Artificial Intelligence and Machine Learning
1. Artificial Intelligence and Machine Learning
MIT OpenCourseWare
40 2: Resistor Capacitor Circuit and Nernst Potential - Intro to Neural Computation
2: Resistor Capacitor Circuit and Nernst Potential - Intro to Neural Computation
MIT OpenCourseWare
41 14: Rate Models and Perceptrons - Intro to Neural Computation
14: Rate Models and Perceptrons - Intro to Neural Computation
MIT OpenCourseWare
42 4: Hodgkin-Huxley Model Part 1 - Intro to Neural Computation
4: Hodgkin-Huxley Model Part 1 - Intro to Neural Computation
MIT OpenCourseWare
43 18: Recurrent Networks - Intro to Neural Computation
18: Recurrent Networks - Intro to Neural Computation
MIT OpenCourseWare
44 3: Resistor Capacitor Neuron Model - Intro to Neural Computation
3: Resistor Capacitor Neuron Model - Intro to Neural Computation
MIT OpenCourseWare
45 15: Matrix Operations - Intro to Neural Computation
15: Matrix Operations - Intro to Neural Computation
MIT OpenCourseWare
46 13: Spectral Analysis Part 3 - Intro to Neural Computation
13: Spectral Analysis Part 3 - Intro to Neural Computation
MIT OpenCourseWare
47 16: Basis Sets - Intro to Neural Computation
16: Basis Sets - Intro to Neural Computation
MIT OpenCourseWare
48 20: Hopfield Networks - Intro to Neural Computation
20: Hopfield Networks - Intro to Neural Computation
MIT OpenCourseWare
49 8: Spike Trains - Intro to Neural Computation
8: Spike Trains - Intro to Neural Computation
MIT OpenCourseWare
50 7: Synapses - Intro to Neural Computation
7: Synapses - Intro to Neural Computation
MIT OpenCourseWare
51 19: Neural Integrators - Intro to Neural Computation
19: Neural Integrators - Intro to Neural Computation
MIT OpenCourseWare
52 5: Hodgkin-Huxley Model Part 2 - Intro to Neural Computation
5: Hodgkin-Huxley Model Part 2 - Intro to Neural Computation
MIT OpenCourseWare
53 6: Dendrites - Intro to Neural Computation
6: Dendrites - Intro to Neural Computation
MIT OpenCourseWare
54 17: Principal Components Analysis_ - Intro to Neural Computation
17: Principal Components Analysis_ - Intro to Neural Computation
MIT OpenCourseWare
55 12: Spectral Analysis Part 2 - Intro to Neural Computation
12: Spectral Analysis Part 2 - Intro to Neural Computation
MIT OpenCourseWare
56 11: Spectral Analysis Part 1 - Intro to Neural Computation
11: Spectral Analysis Part 1 - Intro to Neural Computation
MIT OpenCourseWare
57 9: Receptive Fields - Intro to Neural Computation
9: Receptive Fields - Intro to Neural Computation
MIT OpenCourseWare
58 10: Time Series - Intro to Neural Computation
10: Time Series - Intro to Neural Computation
MIT OpenCourseWare
59 1: Course Overview and Ionic Currents - Intro to Neural Computation
1: Course Overview and Ionic Currents - Intro to Neural Computation
MIT OpenCourseWare
60 The Power of OER with Profs. Mary Rowe and Elizabeth Siler (S1:E10)
The Power of OER with Profs. Mary Rowe and Elizabeth Siler (S1:E10)
MIT OpenCourseWare

Related AI Lessons

H2O.ai launches tabH2O, a foundation model that makes predictions from tabular data without any training
H2O.ai launches tabH2O, a foundation model that makes predictions from tabular data without training, revolutionizing predictive AI for enterprises
The Next Web AI
How TraceML Measures PyTorch Training Time Without Stalling the GPU
Learn how TraceML measures PyTorch training time without stalling the GPU, a crucial technique for optimizing AI model performance
Medium · AI
How TraceML Measures PyTorch Training Time Without Stalling the GPU
Learn how TraceML measures PyTorch training time without stalling the GPU, optimizing ML workflow efficiency
Medium · Machine Learning
How TraceML Measures PyTorch Training Time Without Stalling the GPU
Learn how TraceML measures PyTorch training time without stalling the GPU, optimizing performance for data scientists
Medium · Data Science
Up next
AI Dev 26 x SF: Emma McGrattan: Engineering the Context Layer
DeepLearningAI
Watch →