- published: 01 Feb 2022
- views: 239
-
remove the playlistJudy Hoffman
- remove the playlistJudy Hoffman
Judy Hoffman
Judy Hoffman is an American artist living and working in New York City. Hoffman’s work, which includes installation, sculpture, and artist books, explores themes of birth, decay, waste, and regeneration. Combining disparate elements of industrial refuse, natural debris and handmade paper, she constructs microcosms, ecosystems, and other natural formations.
“Her fascination with making things is evident in the magical way that she combines and constructs disparate elements. Hoffman sees the potential for abject scavenged objects to become something else. It’s as if a crumpled piece of wire calls out to her from the sidewalk, ‘I’m lively. Take me. I could be something,’” wrote Jennifer McGregor, curator at Wave Hill in Bronx, New York.
Hoffman’s environmental installations bring a wildness to manmade spaces. Her site specific installation venues in New York City include Wave Hill, Ceres Gallery, Nutureart, Proteus Gowanus, Kentler International Drawing Center and the Nathan Cummings Foundation; Rutgers and Jersey City Universities, New Jersey; Mary Grove College, Detroit; and the Bienalle Bonn/Frauen Museum and Kunstler Forum in Bonn, Germany.
This page contains text from Wikipedia, the Free Encyclopedia - https://wn.com/Judy_Hoffman
Judy Hoffman (filmmaker)
Judy Hoffman is an American filmmaker and arts activist based in Chicago. She graduated from Northwestern University with a MFA and currently holds the position of Professor of Practice in The Arts in the Department of Cinema and Media Studies at the University of Chicago. Hoffman has played a major role in the development of Kartemquin films working on films such as Golub (2004). Hoffman still continues to have a huge influence in Kartemquin today, serving as a member on the Board of Directors. Beyond Kartemquin films, Hoffman worked with Kwakwaka’wakw, a First Nation in British Columbia. There, she produced films and videotapes about the reclaiming of Native culture.Most of Hoffman's work is done in the form of documentaries focusing on Chicago and its development. Hoffman has brought her activism to her films and still continues to make films that show the many facets of Chicago.
Career
Ms. Hoffman has worked in the film industry for over 35 years; Some of her major work was with the Chicago film company Kartemquin. While at Kartemquin, Hoffman directed Golub, which went on to debut at the New York Film Festival. The mission of Kartemquin, and a guiding theme in Hoffman's work, was the principle of social inquiry; promoting social change through the medium of film. Hoffman still plays a major role at Kartemquin today. Besides working with Kartemquin, Hoffman also worked with the Kwakwaka'wakw First Nation of British Columbia, producing films and videotapes about reclaiming the tribe's culture. Hoffman specifically worked with the video training program on the N'amgis Reserve so that the natives would be able to make their own tapes. While with Kwakwaka'wakw she produced such films as award winning Box of Treasures, which was a film about the efforts to repatriate cultural artifacts.
This page contains text from Wikipedia, the Free Encyclopedia - https://wn.com/Judy_Hoffman_(filmmaker)
Podcasts:
-
Understanding and Mitigating Bias in Vision Systems | Judy Hoffman; Assistant Professor @GT
As visual recognition models are developed across diverse applications; we need the ability to reliably deploy our systems in a variety of environments. At the same time, visual models tend to be trained and evaluated on a static set of curated and annotated data which only represents a subset of the world. In this talk, I will then cover techniques for transferring information between different visual environments and across different semantic tasks thereby enabling recognition models to generalize to previously unseen worlds, such as from simulated to real-world driving imagery. Finally, I'll touch on the pervasiveness of dataset bias and how this bias can adversely affect underrepresented subpopulations.
published: 01 Feb 2022 -
Georgia Tech's Judy Hoffman: The Perils of Learning from Biased Data
Welcome to our seminar series "Applications of Data Science and AI to Equity, Race, and Inclusion in Mobility and Transportation." This topic brings a unique and innovative perspective to existing discussions around diversity, equity, and inclusion. Our aim with these series is to reflect on and raise awareness of applications, opportunities, and potential misuses of these techniques in the mobility and transportation space, specifically as it refers to race, equity, and diversity. The Perils of Learning from Biased Data Abstract: A key task for safely deploying autonomous vehicles is to have reliable perception and understanding of the visual world. Modern computer vision systems can leverage large manually annotated visual datasets to learn visual recognition models that can seemingly r...
published: 24 Apr 2021 -
[OOD Workshop@CoRL 2023] Judy Hoffman - Reliable Vision for a Changing World
Live stream for the first workshop on “Out-of-Distribution Generalization in Robotics: Towards Reliable Learning-Based Autonomy,” held on November 6, 2023 at the 2023 Conference on Robot Learning in Atlanta, Georgia. Workshop site: https://tinyurl.com/corl23ood Invited talk details: Title: "Reliable Vision for a Changing World" Presenter: Prof. Judy Hoffman (Georgia Tech)
published: 05 Dec 2023 -
Judy Hoffman: Responsible CV - Why do models fail and what can we do about it
1st Tutorial on Human-centered AI for Computer Vision at CVPR'22. The slide is available at https://human-centeredai.github.io/
published: 26 Jun 2022 -
[AROW @ ECCV20] Judy Hoffman - Achieving and Understanding Adversarial Robustness
published: 26 Aug 2020 -
Judy Hoffman | CVPR 2020 Embodied AI Workshop Talk
Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech. Previously, she was a Research Scientist at Facebook AI Research. She was a Postdoctoral Researcher at UC Berkeley with Alyosha Efros and Trevor Darrell and at Stanford with Fei-Fei Li. She received my PhD in Electrical Engineering and Computer Science from UC Berkeley in 2016, where she was advised by Trevor Darrell and Kate Saenko. She was awarded the NSF Graduate Fellowship and the Rosalie M. Stern Fellowship. Her thesis focused on transferrable representation learning for visual recognition. For more information about Embodied AI, visit https://embodied-ai.org/
published: 13 Jun 2020 -
Judy Hoffman | On Propaganda and National Identity
Filmmaker and University of Chicago professor Judy Hoffman discusses Triumph of the Will and it’s connection to contemporary propaganda at our Effectiveness of Propaganda Teach-In event.
published: 28 Apr 2017 -
Judy Hoffman (Georgia Tech): ICCV 2021 Workshop on Multi-Task Learning in Computer Vision
Workshop webpage: https://sites.google.com/view/deepmtlworkshop/home
published: 11 Jan 2022 -
Judy Hoffman | On Why We Watch Triumph of the Will
Filmmaker and University of Chicago professor Judy Hoffman discusses Triumph of the Will and it’s connection to contemporary propaganda at our Effectiveness of Propaganda Teach-In event.
published: 28 Apr 2017 -
S2E06: Judy Hoffman with Dhruv Batra on Humans of AI: Stories, Not Stats
Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech and a member of the Machine Learning Center. Find out more about her at https://www.cc.gatech.edu/~judy/. Humans of AI: Stories, Not Stats is an interview series with AI researchers to get to know them better as people. We don't talk about AI or their work or the stats of their life like what college they went to. They share what they think about, what they are insecure about, what they get excited about. They share the stories of their day-to-day life. All interviews in the series are available at http://humanstories.ai. The host of this episode is Dhruv Batra, an Associate Professor in the School of Interactive Computing at Georgia Tech and a Research Scientist at Facebook AI Research (FAIR)....
published: 31 Aug 2021
developed with YouTube
1:01:50
Understanding and Mitigating Bias in Vision Systems | Judy Hoffman; Assistant Professor @GT
- Order: Reorder
- Duration: 1:01:50
- Uploaded Date: 01 Feb 2022
- views: 239
As visual recognition models are developed across diverse applications; we need the ability to reliably deploy our systems in a variety of environments. At the ...
As visual recognition models are developed across diverse applications; we need the ability to reliably deploy our systems in a variety of environments. At the same time, visual models tend to be trained and evaluated on a static set of curated and annotated data which only represents a subset of the world. In this talk, I will then cover techniques for transferring information between different visual environments and across different semantic tasks thereby enabling recognition models to generalize to previously unseen worlds, such as from simulated to real-world driving imagery. Finally, I'll touch on the pervasiveness of dataset bias and how this bias can adversely affect underrepresented subpopulations.
https://wn.com/Understanding_And_Mitigating_Bias_In_Vision_Systems_|_Judy_Hoffman_Assistant_Professor_Gt
As visual recognition models are developed across diverse applications; we need the ability to reliably deploy our systems in a variety of environments. At the same time, visual models tend to be trained and evaluated on a static set of curated and annotated data which only represents a subset of the world. In this talk, I will then cover techniques for transferring information between different visual environments and across different semantic tasks thereby enabling recognition models to generalize to previously unseen worlds, such as from simulated to real-world driving imagery. Finally, I'll touch on the pervasiveness of dataset bias and how this bias can adversely affect underrepresented subpopulations.
59:23
Georgia Tech's Judy Hoffman: The Perils of Learning from Biased Data
- Order: Reorder
- Duration: 59:23
- Uploaded Date: 24 Apr 2021
- views: 235
Welcome to our seminar series "Applications of Data Science and AI to Equity, Race, and Inclusion in Mobility and Transportation." This topic brings a unique an...
Welcome to our seminar series "Applications of Data Science and AI to Equity, Race, and Inclusion in Mobility and Transportation." This topic brings a unique and innovative perspective to existing discussions around diversity, equity, and inclusion. Our aim with these series is to reflect on and raise awareness of applications, opportunities, and potential misuses of these techniques in the mobility and transportation space, specifically as it refers to race, equity, and diversity.
The Perils of Learning from Biased Data
Abstract:
A key task for safely deploying autonomous vehicles is to have reliable perception and understanding of the visual world. Modern computer vision systems can leverage large manually annotated visual datasets to learn visual recognition models that can seemingly replicate the behavior of the human annotators, automatically recognizing cars on the road and pedestrians crossing the street. In this talk, I will discuss a central challenge with learning-based systems which is that they rely on sufficiently diverse training data to capture all data variance anticipated at deployment. When the data used for learning in fact only represents a biased subset of the world, the model may suffer poor predictive performance when the test time bias changes. I will explore how this situation arises under benign conditions, such as weather pattern changes, as well as how systematic demographic bias can lead to inequitable predictive performance across subpopulations.
Judy Hoffman
Dr. Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech and a member of the Machine Learning Center. Her research lies at the intersection of computer vision and machine learning with specialization in domain adaptation, transfer learning, adversarial robustness, and algorithmic fairness. She has been awarded the NVIDIA female leader in computer vision award in 2020, AIMiner top 100 most influential scholars in Machine Learning (2020), MIT EECS Rising Star in 2015, and is a recipient of the NSF Graduate Fellowship. In addition to her research, she co-founded and continues to advise for Women in Computer Vision, an organization which provides mentorship and travel support for early-career women in the computer vision community. Prior to joining Georgia Tech, she was a Research Scientist at Facebook AI Research. She received her PhD in Electrical Engineering and Computer Science from UC Berkeley in 2016 after which she completed Postdocs at Stanford University (2017) and UC Berkeley (2018).
https://wn.com/Georgia_Tech's_Judy_Hoffman_The_Perils_Of_Learning_From_Biased_Data
Welcome to our seminar series "Applications of Data Science and AI to Equity, Race, and Inclusion in Mobility and Transportation." This topic brings a unique and innovative perspective to existing discussions around diversity, equity, and inclusion. Our aim with these series is to reflect on and raise awareness of applications, opportunities, and potential misuses of these techniques in the mobility and transportation space, specifically as it refers to race, equity, and diversity.
The Perils of Learning from Biased Data
Abstract:
A key task for safely deploying autonomous vehicles is to have reliable perception and understanding of the visual world. Modern computer vision systems can leverage large manually annotated visual datasets to learn visual recognition models that can seemingly replicate the behavior of the human annotators, automatically recognizing cars on the road and pedestrians crossing the street. In this talk, I will discuss a central challenge with learning-based systems which is that they rely on sufficiently diverse training data to capture all data variance anticipated at deployment. When the data used for learning in fact only represents a biased subset of the world, the model may suffer poor predictive performance when the test time bias changes. I will explore how this situation arises under benign conditions, such as weather pattern changes, as well as how systematic demographic bias can lead to inequitable predictive performance across subpopulations.
Judy Hoffman
Dr. Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech and a member of the Machine Learning Center. Her research lies at the intersection of computer vision and machine learning with specialization in domain adaptation, transfer learning, adversarial robustness, and algorithmic fairness. She has been awarded the NVIDIA female leader in computer vision award in 2020, AIMiner top 100 most influential scholars in Machine Learning (2020), MIT EECS Rising Star in 2015, and is a recipient of the NSF Graduate Fellowship. In addition to her research, she co-founded and continues to advise for Women in Computer Vision, an organization which provides mentorship and travel support for early-career women in the computer vision community. Prior to joining Georgia Tech, she was a Research Scientist at Facebook AI Research. She received her PhD in Electrical Engineering and Computer Science from UC Berkeley in 2016 after which she completed Postdocs at Stanford University (2017) and UC Berkeley (2018).
- published: 24 Apr 2021
- views: 235
32:13
[OOD Workshop@CoRL 2023] Judy Hoffman - Reliable Vision for a Changing World
- Order: Reorder
- Duration: 32:13
- Uploaded Date: 05 Dec 2023
- views: 70
Live stream for the first workshop on “Out-of-Distribution Generalization in Robotics: Towards Reliable Learning-Based Autonomy,” held on November 6, 2023 at th...
Live stream for the first workshop on “Out-of-Distribution Generalization in Robotics: Towards Reliable Learning-Based Autonomy,” held on November 6, 2023 at the 2023 Conference on Robot Learning in Atlanta, Georgia.
Workshop site: https://tinyurl.com/corl23ood
Invited talk details:
Title: "Reliable Vision for a Changing World"
Presenter: Prof. Judy Hoffman (Georgia Tech)
https://wn.com/Ood_Workshop_Corl_2023_Judy_Hoffman_Reliable_Vision_For_A_Changing_World
Live stream for the first workshop on “Out-of-Distribution Generalization in Robotics: Towards Reliable Learning-Based Autonomy,” held on November 6, 2023 at the 2023 Conference on Robot Learning in Atlanta, Georgia.
Workshop site: https://tinyurl.com/corl23ood
Invited talk details:
Title: "Reliable Vision for a Changing World"
Presenter: Prof. Judy Hoffman (Georgia Tech)
- published: 05 Dec 2023
- views: 70
54:32
Judy Hoffman: Responsible CV - Why do models fail and what can we do about it
- Order: Reorder
- Duration: 54:32
- Uploaded Date: 26 Jun 2022
- views: 349
1st Tutorial on Human-centered AI for Computer Vision at CVPR'22.
The slide is available at https://human-centeredai.github.io/
1st Tutorial on Human-centered AI for Computer Vision at CVPR'22.
The slide is available at https://human-centeredai.github.io/
https://wn.com/Judy_Hoffman_Responsible_Cv_Why_Do_Models_Fail_And_What_Can_We_Do_About_It
1st Tutorial on Human-centered AI for Computer Vision at CVPR'22.
The slide is available at https://human-centeredai.github.io/
- published: 26 Jun 2022
- views: 349
27:42
[AROW @ ECCV20] Judy Hoffman - Achieving and Understanding Adversarial Robustness
- Order: Reorder
- Duration: 27:42
- Uploaded Date: 26 Aug 2020
- views: 206
- published: 26 Aug 2020
- views: 206
28:27
Judy Hoffman | CVPR 2020 Embodied AI Workshop Talk
- Order: Reorder
- Duration: 28:27
- Uploaded Date: 13 Jun 2020
- views: 1230
Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech. Previously, she was a Research Scientist at Facebook AI Research....
Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech. Previously, she was a Research Scientist at Facebook AI Research. She was a Postdoctoral Researcher at UC Berkeley with Alyosha Efros and Trevor Darrell and at Stanford with Fei-Fei Li. She received my PhD in Electrical Engineering and Computer Science from UC Berkeley in 2016, where she was advised by Trevor Darrell and Kate Saenko. She was awarded the NSF Graduate Fellowship and the Rosalie M. Stern Fellowship. Her thesis focused on transferrable representation learning for visual recognition.
For more information about Embodied AI, visit
https://embodied-ai.org/
https://wn.com/Judy_Hoffman_|_Cvpr_2020_Embodied_Ai_Workshop_Talk
Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech. Previously, she was a Research Scientist at Facebook AI Research. She was a Postdoctoral Researcher at UC Berkeley with Alyosha Efros and Trevor Darrell and at Stanford with Fei-Fei Li. She received my PhD in Electrical Engineering and Computer Science from UC Berkeley in 2016, where she was advised by Trevor Darrell and Kate Saenko. She was awarded the NSF Graduate Fellowship and the Rosalie M. Stern Fellowship. Her thesis focused on transferrable representation learning for visual recognition.
For more information about Embodied AI, visit
https://embodied-ai.org/
- published: 13 Jun 2020
- views: 1230
1:30
Judy Hoffman | On Propaganda and National Identity
- Order: Reorder
- Duration: 1:30
- Uploaded Date: 28 Apr 2017
- views: 52
Filmmaker and University of Chicago professor Judy Hoffman discusses Triumph of the Will and it’s connection to contemporary propaganda at our Effectiveness of ...
Filmmaker and University of Chicago professor Judy Hoffman discusses Triumph of the Will and it’s connection to contemporary propaganda at our Effectiveness of Propaganda Teach-In event.
https://wn.com/Judy_Hoffman_|_On_Propaganda_And_National_Identity
Filmmaker and University of Chicago professor Judy Hoffman discusses Triumph of the Will and it’s connection to contemporary propaganda at our Effectiveness of Propaganda Teach-In event.
- published: 28 Apr 2017
- views: 52
39:46
Judy Hoffman (Georgia Tech): ICCV 2021 Workshop on Multi-Task Learning in Computer Vision
- Order: Reorder
- Duration: 39:46
- Uploaded Date: 11 Jan 2022
- views: 230
Workshop webpage: https://sites.google.com/view/deepmtlworkshop/home
Workshop webpage: https://sites.google.com/view/deepmtlworkshop/home
https://wn.com/Judy_Hoffman_(Georgia_Tech)_Iccv_2021_Workshop_On_Multi_Task_Learning_In_Computer_Vision
Workshop webpage: https://sites.google.com/view/deepmtlworkshop/home
- published: 11 Jan 2022
- views: 230
2:18
Judy Hoffman | On Why We Watch Triumph of the Will
- Order: Reorder
- Duration: 2:18
- Uploaded Date: 28 Apr 2017
- views: 692
Filmmaker and University of Chicago professor Judy Hoffman discusses Triumph of the Will and it’s connection to contemporary propaganda at our Effectiveness of ...
Filmmaker and University of Chicago professor Judy Hoffman discusses Triumph of the Will and it’s connection to contemporary propaganda at our Effectiveness of Propaganda Teach-In event.
https://wn.com/Judy_Hoffman_|_On_Why_We_Watch_Triumph_Of_The_Will
Filmmaker and University of Chicago professor Judy Hoffman discusses Triumph of the Will and it’s connection to contemporary propaganda at our Effectiveness of Propaganda Teach-In event.
- published: 28 Apr 2017
- views: 692
51:46
S2E06: Judy Hoffman with Dhruv Batra on Humans of AI: Stories, Not Stats
- Order: Reorder
- Duration: 51:46
- Uploaded Date: 31 Aug 2021
- views: 1470
Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech and a member of the Machine Learning Center. Find out more about h...
Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech and a member of the Machine Learning Center. Find out more about her at https://www.cc.gatech.edu/~judy/.
Humans of AI: Stories, Not Stats is an interview series with AI researchers to get to know them better as people. We don't talk about AI or their work or the stats of their life like what college they went to. They share what they think about, what they are insecure about, what they get excited about. They share the stories of their day-to-day life.
All interviews in the series are available at http://humanstories.ai.
The host of this episode is Dhruv Batra, an Associate Professor in the School of Interactive Computing at Georgia Tech and a Research Scientist at Facebook AI Research (FAIR). Find out more about him at https://www.cc.gatech.edu/~dbatra/ or follow him on Twitter https://twitter.com/DhruvBatraDB.
This interview was recorded on 26th March 2021.
This interview is available as a podcast episode at https://anchor.fm/humanstoriesai/episodes/S2E6-Judy-Hoffman-with-Dhruv-Batra-e14rokb.
Edits and timeline annotations thanks to Mukul Khanna https://twitter.com/mkulkhanna and Varshini Subhash https://twitter.com/VarshiniSubhash.
00:00 Opening
00:49 Introduction
01:30 What were you doing just before this call?
01:43 What is your daily routine like?
02:21 What is the favorite part of your day?
02:53 What is the least favorite part of your day?
03:41 Do you set an alarm in the morning?
03:55 Do you hit the snooze button?
04:36 Are you a planner or do you operate on gut-feeling?
05:32 Do you struggle with procrastination?
06:29 Do you struggle with time management?
06:57 Are you competitive?
08:06 Is there a rejection or a failure that hurt particularly bad?
09:26 Is there an achievement or a success that felt particularly good?
10:35 What is one thing you are worse at than people around you?
11:34 What is your single biggest strength?
12:13 What is your one favorite tool/trick/hack that makes your life more convenient or efficient or fun?
13:47 What is an idea or a book or essay or movie or podcasts or external influences of any sort that left a particularly deep impression on you?
15:01 How do you usually make difficult decisions? Are there certain lines of thinking or mental frameworks you use?
15:34 Do you find that you are happy with these decisions? Do you tend to have regrets later?
16:47 Do you have an internal monologue? Do you talk to yourself? If yes, in what language?
17:20 Are you a visual thinker?
18:29 What do you tend to think about most when you are not intentionally trying to think about something?
19:22 How do you recharge or take a break?
20:09 Where do you find your escape?
21:00 Are you happy with the number of close friends you have?
21:35 What are you insecure about?
22:27 Do you think you are average, above average, or below average happy than people around you?
23:46 What is something surprising about you? Something the rest of us might not guess.
25:24 What is one thing about the world that surprises you?
28:02 What do you wish your brain was better at doing?
28:54 What do you strongly suspect but have no proof of?
29:55 What is something you've changed your mind about?
32:22 What is a bad habit you're working on overcoming?
33:48 What are you addicted to?
34:40 How do you imagine your retirement?
36:05 What age are you imagining when thinking about retirement?
36:34 Do you think about the future much (say on a 5-10 year scale)?
37:15 When do you think the world will open back up?
38:39 Do you think there is a point to life, our existence?
41:43 What do you find meaning in?
42:39 Pineapple on pizza? Yummy or an abomination?
43:06 How do you decide what to work on?
44:28 How do you capture and keep track of ideas?
45:51 What are some traits common across some of the best collaborators/colleagues you've worked with?
47:22 How do you spot these traits early? Are you good at spotting them?
48:10 Describe something that has made you smile today.
48:27 What is some of the best advice you've gotten or given?
50:15 Why did you agree to do this interview with me?
51:10 Is there anything you'd like to talk about in terms of who you are, what your life is like, that we didn't cover?
https://wn.com/S2E06_Judy_Hoffman_With_Dhruv_Batra_On_Humans_Of_Ai_Stories,_Not_Stats
Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech and a member of the Machine Learning Center. Find out more about her at https://www.cc.gatech.edu/~judy/.
Humans of AI: Stories, Not Stats is an interview series with AI researchers to get to know them better as people. We don't talk about AI or their work or the stats of their life like what college they went to. They share what they think about, what they are insecure about, what they get excited about. They share the stories of their day-to-day life.
All interviews in the series are available at http://humanstories.ai.
The host of this episode is Dhruv Batra, an Associate Professor in the School of Interactive Computing at Georgia Tech and a Research Scientist at Facebook AI Research (FAIR). Find out more about him at https://www.cc.gatech.edu/~dbatra/ or follow him on Twitter https://twitter.com/DhruvBatraDB.
This interview was recorded on 26th March 2021.
This interview is available as a podcast episode at https://anchor.fm/humanstoriesai/episodes/S2E6-Judy-Hoffman-with-Dhruv-Batra-e14rokb.
Edits and timeline annotations thanks to Mukul Khanna https://twitter.com/mkulkhanna and Varshini Subhash https://twitter.com/VarshiniSubhash.
00:00 Opening
00:49 Introduction
01:30 What were you doing just before this call?
01:43 What is your daily routine like?
02:21 What is the favorite part of your day?
02:53 What is the least favorite part of your day?
03:41 Do you set an alarm in the morning?
03:55 Do you hit the snooze button?
04:36 Are you a planner or do you operate on gut-feeling?
05:32 Do you struggle with procrastination?
06:29 Do you struggle with time management?
06:57 Are you competitive?
08:06 Is there a rejection or a failure that hurt particularly bad?
09:26 Is there an achievement or a success that felt particularly good?
10:35 What is one thing you are worse at than people around you?
11:34 What is your single biggest strength?
12:13 What is your one favorite tool/trick/hack that makes your life more convenient or efficient or fun?
13:47 What is an idea or a book or essay or movie or podcasts or external influences of any sort that left a particularly deep impression on you?
15:01 How do you usually make difficult decisions? Are there certain lines of thinking or mental frameworks you use?
15:34 Do you find that you are happy with these decisions? Do you tend to have regrets later?
16:47 Do you have an internal monologue? Do you talk to yourself? If yes, in what language?
17:20 Are you a visual thinker?
18:29 What do you tend to think about most when you are not intentionally trying to think about something?
19:22 How do you recharge or take a break?
20:09 Where do you find your escape?
21:00 Are you happy with the number of close friends you have?
21:35 What are you insecure about?
22:27 Do you think you are average, above average, or below average happy than people around you?
23:46 What is something surprising about you? Something the rest of us might not guess.
25:24 What is one thing about the world that surprises you?
28:02 What do you wish your brain was better at doing?
28:54 What do you strongly suspect but have no proof of?
29:55 What is something you've changed your mind about?
32:22 What is a bad habit you're working on overcoming?
33:48 What are you addicted to?
34:40 How do you imagine your retirement?
36:05 What age are you imagining when thinking about retirement?
36:34 Do you think about the future much (say on a 5-10 year scale)?
37:15 When do you think the world will open back up?
38:39 Do you think there is a point to life, our existence?
41:43 What do you find meaning in?
42:39 Pineapple on pizza? Yummy or an abomination?
43:06 How do you decide what to work on?
44:28 How do you capture and keep track of ideas?
45:51 What are some traits common across some of the best collaborators/colleagues you've worked with?
47:22 How do you spot these traits early? Are you good at spotting them?
48:10 Describe something that has made you smile today.
48:27 What is some of the best advice you've gotten or given?
50:15 Why did you agree to do this interview with me?
51:10 Is there anything you'd like to talk about in terms of who you are, what your life is like, that we didn't cover?
- published: 31 Aug 2021
- views: 1470
1:01:50
Understanding and Mitigating Bias in Vision Systems | Judy Hoffman; Assistant Professor @GT
As visual recognition models are developed across diverse applications; we need the abilit...
published: 01 Feb 2022
Understanding and Mitigating Bias in Vision Systems | Judy Hoffman; Assistant Professor @GT
Understanding and Mitigating Bias in Vision Systems | Judy Hoffman; Assistant Professor @GT
- Report rights infringement
- published: 01 Feb 2022
- views: 239
As visual recognition models are developed across diverse applications; we need the ability to reliably deploy our systems in a variety of environments. At the same time, visual models tend to be trained and evaluated on a static set of curated and annotated data which only represents a subset of the world. In this talk, I will then cover techniques for transferring information between different visual environments and across different semantic tasks thereby enabling recognition models to generalize to previously unseen worlds, such as from simulated to real-world driving imagery. Finally, I'll touch on the pervasiveness of dataset bias and how this bias can adversely affect underrepresented subpopulations.
59:23
Georgia Tech's Judy Hoffman: The Perils of Learning from Biased Data
Welcome to our seminar series "Applications of Data Science and AI to Equity, Race, and In...
published: 24 Apr 2021
Georgia Tech's Judy Hoffman: The Perils of Learning from Biased Data
Georgia Tech's Judy Hoffman: The Perils of Learning from Biased Data
- Report rights infringement
- published: 24 Apr 2021
- views: 235
Welcome to our seminar series "Applications of Data Science and AI to Equity, Race, and Inclusion in Mobility and Transportation." This topic brings a unique and innovative perspective to existing discussions around diversity, equity, and inclusion. Our aim with these series is to reflect on and raise awareness of applications, opportunities, and potential misuses of these techniques in the mobility and transportation space, specifically as it refers to race, equity, and diversity.
The Perils of Learning from Biased Data
Abstract:
A key task for safely deploying autonomous vehicles is to have reliable perception and understanding of the visual world. Modern computer vision systems can leverage large manually annotated visual datasets to learn visual recognition models that can seemingly replicate the behavior of the human annotators, automatically recognizing cars on the road and pedestrians crossing the street. In this talk, I will discuss a central challenge with learning-based systems which is that they rely on sufficiently diverse training data to capture all data variance anticipated at deployment. When the data used for learning in fact only represents a biased subset of the world, the model may suffer poor predictive performance when the test time bias changes. I will explore how this situation arises under benign conditions, such as weather pattern changes, as well as how systematic demographic bias can lead to inequitable predictive performance across subpopulations.
Judy Hoffman
Dr. Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech and a member of the Machine Learning Center. Her research lies at the intersection of computer vision and machine learning with specialization in domain adaptation, transfer learning, adversarial robustness, and algorithmic fairness. She has been awarded the NVIDIA female leader in computer vision award in 2020, AIMiner top 100 most influential scholars in Machine Learning (2020), MIT EECS Rising Star in 2015, and is a recipient of the NSF Graduate Fellowship. In addition to her research, she co-founded and continues to advise for Women in Computer Vision, an organization which provides mentorship and travel support for early-career women in the computer vision community. Prior to joining Georgia Tech, she was a Research Scientist at Facebook AI Research. She received her PhD in Electrical Engineering and Computer Science from UC Berkeley in 2016 after which she completed Postdocs at Stanford University (2017) and UC Berkeley (2018).
32:13
[OOD Workshop@CoRL 2023] Judy Hoffman - Reliable Vision for a Changing World
Live stream for the first workshop on “Out-of-Distribution Generalization in Robotics: Tow...
published: 05 Dec 2023
[OOD Workshop@CoRL 2023] Judy Hoffman - Reliable Vision for a Changing World
[OOD Workshop@CoRL 2023] Judy Hoffman - Reliable Vision for a Changing World
- Report rights infringement
- published: 05 Dec 2023
- views: 70
Live stream for the first workshop on “Out-of-Distribution Generalization in Robotics: Towards Reliable Learning-Based Autonomy,” held on November 6, 2023 at the 2023 Conference on Robot Learning in Atlanta, Georgia.
Workshop site: https://tinyurl.com/corl23ood
Invited talk details:
Title: "Reliable Vision for a Changing World"
Presenter: Prof. Judy Hoffman (Georgia Tech)
54:32
Judy Hoffman: Responsible CV - Why do models fail and what can we do about it
1st Tutorial on Human-centered AI for Computer Vision at CVPR'22.
The slide is available a...
published: 26 Jun 2022
Judy Hoffman: Responsible CV - Why do models fail and what can we do about it
Judy Hoffman: Responsible CV - Why do models fail and what can we do about it
- Report rights infringement
- published: 26 Jun 2022
- views: 349
1st Tutorial on Human-centered AI for Computer Vision at CVPR'22.
The slide is available at https://human-centeredai.github.io/
27:42
[AROW @ ECCV20] Judy Hoffman - Achieving and Understanding Adversarial Robustness
published: 26 Aug 2020
[AROW @ ECCV20] Judy Hoffman - Achieving and Understanding Adversarial Robustness
[AROW @ ECCV20] Judy Hoffman - Achieving and Understanding Adversarial Robustness
- Report rights infringement
- published: 26 Aug 2020
- views: 206
28:27
Judy Hoffman | CVPR 2020 Embodied AI Workshop Talk
Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia T...
published: 13 Jun 2020
Judy Hoffman | CVPR 2020 Embodied AI Workshop Talk
Judy Hoffman | CVPR 2020 Embodied AI Workshop Talk
- Report rights infringement
- published: 13 Jun 2020
- views: 1230
Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech. Previously, she was a Research Scientist at Facebook AI Research. She was a Postdoctoral Researcher at UC Berkeley with Alyosha Efros and Trevor Darrell and at Stanford with Fei-Fei Li. She received my PhD in Electrical Engineering and Computer Science from UC Berkeley in 2016, where she was advised by Trevor Darrell and Kate Saenko. She was awarded the NSF Graduate Fellowship and the Rosalie M. Stern Fellowship. Her thesis focused on transferrable representation learning for visual recognition.
For more information about Embodied AI, visit
https://embodied-ai.org/
1:30
Judy Hoffman | On Propaganda and National Identity
Filmmaker and University of Chicago professor Judy Hoffman discusses Triumph of the Will a...
published: 28 Apr 2017
Judy Hoffman | On Propaganda and National Identity
Judy Hoffman | On Propaganda and National Identity
- Report rights infringement
- published: 28 Apr 2017
- views: 52
Filmmaker and University of Chicago professor Judy Hoffman discusses Triumph of the Will and it’s connection to contemporary propaganda at our Effectiveness of Propaganda Teach-In event.
39:46
Judy Hoffman (Georgia Tech): ICCV 2021 Workshop on Multi-Task Learning in Computer Vision
Workshop webpage: https://sites.google.com/view/deepmtlworkshop/home
published: 11 Jan 2022
Judy Hoffman (Georgia Tech): ICCV 2021 Workshop on Multi-Task Learning in Computer Vision
Judy Hoffman (Georgia Tech): ICCV 2021 Workshop on Multi-Task Learning in Computer Vision
- Report rights infringement
- published: 11 Jan 2022
- views: 230
Workshop webpage: https://sites.google.com/view/deepmtlworkshop/home
2:18
Judy Hoffman | On Why We Watch Triumph of the Will
Filmmaker and University of Chicago professor Judy Hoffman discusses Triumph of the Will a...
published: 28 Apr 2017
Judy Hoffman | On Why We Watch Triumph of the Will
Judy Hoffman | On Why We Watch Triumph of the Will
- Report rights infringement
- published: 28 Apr 2017
- views: 692
Filmmaker and University of Chicago professor Judy Hoffman discusses Triumph of the Will and it’s connection to contemporary propaganda at our Effectiveness of Propaganda Teach-In event.
51:46
S2E06: Judy Hoffman with Dhruv Batra on Humans of AI: Stories, Not Stats
Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia T...
published: 31 Aug 2021
S2E06: Judy Hoffman with Dhruv Batra on Humans of AI: Stories, Not Stats
S2E06: Judy Hoffman with Dhruv Batra on Humans of AI: Stories, Not Stats
- Report rights infringement
- published: 31 Aug 2021
- views: 1470
Judy Hoffman is an Assistant Professor in the School of Interactive Computing at Georgia Tech and a member of the Machine Learning Center. Find out more about her at https://www.cc.gatech.edu/~judy/.
Humans of AI: Stories, Not Stats is an interview series with AI researchers to get to know them better as people. We don't talk about AI or their work or the stats of their life like what college they went to. They share what they think about, what they are insecure about, what they get excited about. They share the stories of their day-to-day life.
All interviews in the series are available at http://humanstories.ai.
The host of this episode is Dhruv Batra, an Associate Professor in the School of Interactive Computing at Georgia Tech and a Research Scientist at Facebook AI Research (FAIR). Find out more about him at https://www.cc.gatech.edu/~dbatra/ or follow him on Twitter https://twitter.com/DhruvBatraDB.
This interview was recorded on 26th March 2021.
This interview is available as a podcast episode at https://anchor.fm/humanstoriesai/episodes/S2E6-Judy-Hoffman-with-Dhruv-Batra-e14rokb.
Edits and timeline annotations thanks to Mukul Khanna https://twitter.com/mkulkhanna and Varshini Subhash https://twitter.com/VarshiniSubhash.
00:00 Opening
00:49 Introduction
01:30 What were you doing just before this call?
01:43 What is your daily routine like?
02:21 What is the favorite part of your day?
02:53 What is the least favorite part of your day?
03:41 Do you set an alarm in the morning?
03:55 Do you hit the snooze button?
04:36 Are you a planner or do you operate on gut-feeling?
05:32 Do you struggle with procrastination?
06:29 Do you struggle with time management?
06:57 Are you competitive?
08:06 Is there a rejection or a failure that hurt particularly bad?
09:26 Is there an achievement or a success that felt particularly good?
10:35 What is one thing you are worse at than people around you?
11:34 What is your single biggest strength?
12:13 What is your one favorite tool/trick/hack that makes your life more convenient or efficient or fun?
13:47 What is an idea or a book or essay or movie or podcasts or external influences of any sort that left a particularly deep impression on you?
15:01 How do you usually make difficult decisions? Are there certain lines of thinking or mental frameworks you use?
15:34 Do you find that you are happy with these decisions? Do you tend to have regrets later?
16:47 Do you have an internal monologue? Do you talk to yourself? If yes, in what language?
17:20 Are you a visual thinker?
18:29 What do you tend to think about most when you are not intentionally trying to think about something?
19:22 How do you recharge or take a break?
20:09 Where do you find your escape?
21:00 Are you happy with the number of close friends you have?
21:35 What are you insecure about?
22:27 Do you think you are average, above average, or below average happy than people around you?
23:46 What is something surprising about you? Something the rest of us might not guess.
25:24 What is one thing about the world that surprises you?
28:02 What do you wish your brain was better at doing?
28:54 What do you strongly suspect but have no proof of?
29:55 What is something you've changed your mind about?
32:22 What is a bad habit you're working on overcoming?
33:48 What are you addicted to?
34:40 How do you imagine your retirement?
36:05 What age are you imagining when thinking about retirement?
36:34 Do you think about the future much (say on a 5-10 year scale)?
37:15 When do you think the world will open back up?
38:39 Do you think there is a point to life, our existence?
41:43 What do you find meaning in?
42:39 Pineapple on pizza? Yummy or an abomination?
43:06 How do you decide what to work on?
44:28 How do you capture and keep track of ideas?
45:51 What are some traits common across some of the best collaborators/colleagues you've worked with?
47:22 How do you spot these traits early? Are you good at spotting them?
48:10 Describe something that has made you smile today.
48:27 What is some of the best advice you've gotten or given?
50:15 Why did you agree to do this interview with me?
51:10 Is there anything you'd like to talk about in terms of who you are, what your life is like, that we didn't cover?
Judy Hoffman
Judy Hoffman is an American artist living and working in New York City. Hoffman’s work, which includes installation, sculpture, and artist books, explores themes of birth, decay, waste, and regeneration. Combining disparate elements of industrial refuse, natural debris and handmade paper, she constructs microcosms, ecosystems, and other natural formations.
“Her fascination with making things is evident in the magical way that she combines and constructs disparate elements. Hoffman sees the potential for abject scavenged objects to become something else. It’s as if a crumpled piece of wire calls out to her from the sidewalk, ‘I’m lively. Take me. I could be something,’” wrote Jennifer McGregor, curator at Wave Hill in Bronx, New York.
Hoffman’s environmental installations bring a wildness to manmade spaces. Her site specific installation venues in New York City include Wave Hill, Ceres Gallery, Nutureart, Proteus Gowanus, Kentler International Drawing Center and the Nathan Cummings Foundation; Rutgers and Jersey City Universities, New Jersey; Mary Grove College, Detroit; and the Bienalle Bonn/Frauen Museum and Kunstler Forum in Bonn, Germany.
This page contains text from Wikipedia, the Free Encyclopedia - https://wn.com/Judy_Hoffman
Understanding and Mitigating Bias in Vision System...
Georgia Tech's Judy Hoffman: The Perils of Learnin...
[OOD Workshop@CoRL 2023] Judy Hoffman - Reliable V...
Judy Hoffman: Responsible CV - Why do models fail ...
[AROW @ ECCV20] Judy Hoffman - Achieving and Under...
Judy Hoffman | CVPR 2020 Embodied AI Workshop Talk...
Judy Hoffman | On Propaganda and National Identity...
Judy Hoffman (Georgia Tech): ICCV 2021 Workshop on...
Judy Hoffman | On Why We Watch Triumph of the Will...