Computer Vision is Not Perfect

FromData Skeptic

Start listening View podcast show

Computer Vision is Not Perfect

FromData Skeptic

ratings:

Length:

26 minutes

Released:

Apr 10, 2020

Format:

Podcast episode

Description

Computer Vision is not Perfect Julia Evans joins us help answer the question why do neural networks think a panda is a vulture. Kyle talks to Julia about her hands-on work fooling neural networks. Julia runs Wizard Zines which publishes works such as Your Linux Toolbox. You can find her on Twitter @b0rk

Released:

Apr 10, 2020

Format:

Podcast episode

Titles in the series (100)

Data Skeptic is a data science podcast exploring machine learning, statistics, artificial intelligence, and other data topics through short tutorials and interviews with domain experts.

Skip carousel

Guerilla Skepticism on Wikipedia with Susan Gerbic: Our guest this week is Susan Gerbic. Susan is a skeptical activist involved in many activities, the one we focus on most in this episode is Guerrilla Skepticism on Wikipedia, an organization working to improve the content and citations of... by Data Skeptic
70 min listen
[MINI] Decision Tree Learning: Linhda and Kyle talk about Decision Tree Learning in this miniepisode. Decision Tree Learning is the algorithmic process of trying to generate an optimal decision tree to properly classify or forecast some future unlabeled element based by... by Data Skeptic
13 min listen
[MINI] Experimental Design: This episode loosely explores the topic of Experimental Design including hypothesis testing, the importance of statistical tests, and an everyday and business example. by Data Skeptic
16 min listen
[MINI] p-values: In this mini, we discuss p-values and their use in hypothesis testing, in the context of an hypothetical experiment on plant flowering, and end with a reference to the Particle Fever documentary and how statistical significance played a role. by Data Skeptic
17 min listen
Introduction: The Data Skeptic Podcast features conversations with topics related to data science, statistics, machine learning, artificial intelligence and the like, all from the perspective of applying critical thinking and the scientific method to evaluate the... by Data Skeptic
4 min listen
[MINI] type i / type ii errors: In this first mini-episode of the Data Skeptic Podcast, we define and discuss type i and type ii errors (a.k.a. false positives and false negatives). by Data Skeptic
11 min listen
The Right (big data) Tool for the Job with Jay Shankar: In this week's episode, we discuss applied solutions to big data problem with big data engineer Jay Shankar. The episode explores approaches and design philosophy to solving real world big data business problems, and the exploration of the wide... by Data Skeptic
50 min listen
Data Myths with Karl Mamer: Black out baby boom | Subliminal messages | Beer and Diapers by Data Skeptic
48 min listen
Data in Healthcare IT with Shahid Shah: Our guest this week is Shahid Shah. Shahid is CEO at Netspective, and writes three blogs: Health Care Guy, Shahid Shah, and HitSphere - the Healthcare IT Supersite. During the program, Kyle recommended a talk from the 2014 MIT Sloan CIO Symposium... by Data Skeptic
57 min listen
Advertising Attribution with Nathan Janos: A conversation with Convertro's Nathan Janos about methodologies used to help advertisers understand the affect each of their marketing efforts (print, SEM, display, skywriting, etc.) contributes to their overall return. by Data Skeptic
76 min listen
Streetlight Outage and Crime Rate Analysis with Zach Seeskin: This episode features a discussion with statistics PhD student Zach Seeskin about a project he was involved in as part of the Eric and Wendy Schmidt Data Science for Social Good Summer Fellowship. The project involved exploring the... by Data Skeptic
33 min listen
[MINI] Ant Colony Optimization: In this week's mini episode, Linhda and Kyle discuss Ant Colony Optimization - a numerical / stochastic optimization technique which models its search after the process ants employ in using random walks to find a goal (food) and then leaving a... by Data Skeptic
15 min listen
Personalized Medicine with Niki Athanasiadou: In the second full length episode of the podcast, we discuss the current state of personalized medicine and the advancements in genetics that have made it possible. by Data Skeptic
57 min listen
Economic Modeling and Prediction, Charitable Giving, and a Follow Up with Peter Backus: Economist Peter Backus joins me in this episode to discuss a few interesting topics. You may recall Linhda and I previously discussed his paper "The Girlfriend Equation" on a recent mini-episode. We start by touching base on this fun paper and get a... by Data Skeptic
24 min listen
Easily Fooling Deep Neural Networks: My guest this week is Anh Nguyen, a PhD student at the University of Wyoming working in the Evolving AI lab. The episode discusses the paper Deep Neural Networks are Easily Fooled [pdf] by Anh Nguyen, Jason Yosinski, and Jeff Clune. It... by Data Skeptic
28 min listen
[MINI] Cross Validation: This miniepisode discusses the technique called Cross Validation - a process by which one randomly divides up a dataset into numerous small partitions. Next, (typically) one is held out, and the rest are used to train some model. The hold out set can... by Data Skeptic
1 min listen
[MINI] The T-Test: The t-test is this week's mini-episode topic. The t-test is a statistical testing procedure used to determine if the mean of two datasets differs by a statistically significant amount. We discuss how a wine manufacturer might apply a t-test to... by Data Skeptic
17 min listen
NYC Speed Camera Analysis with Tim Schmeier: New York State approved the use of automated speed cameras within a specific range of schools. Tim Schmeier did an analysis of publically available data related to these cameras as part of a project at the NYC Data Science Academy. Tim's work... by Data Skeptic
17 min listen
[MINI] Value of Information: A discussion about getting ready in the morning, negotiating a used car purchase, and selecting the best AirBnB place to stay at help frame a conversation about the decision theoretic principal known as the Value of Information equation. by Data Skeptic
14 min listen
[MINI] Bayesian Updating: In this minisode, we discuss Bayesian Updating - the process by which one can calculate the most likely hypothesis might be true given one's older / prior belief and all new evidence. by Data Skeptic
11 min listen
[MINI] Noise!!: Our topic for this week is "noise" as in signal vs. noise. This is not a signal processing discussions, but rather a brief introduction to how the work noise is used to describe how much information in a dataset is useless (as opposed to... by Data Skeptic
16 min listen
[MINI] The Girlfriend Equation: Economist Peter Backus put forward "The Girlfriend Equation" while working on his PhD - a probabilistic model attempting to estimate the likelihood of him finding a girlfriend. In this mini episode we explore the soundness of his model and also share... by Data Skeptic
16 min listen
[MINI] The Battle of the Sexes: Love and Data is the continued theme in this mini-episode as we discuss the game theory example of The Battle of the Sexes. In this textbook example, a couple must strategize about how to spend their Friday night. One partner prefers football games... by Data Skeptic
18 min listen
Doubtful News, Geology, Investigating Paranormal Groups, and Thinking Scientifically with Sharon Hill: I had the change to speak with well known Sharon Hill (@idoubtit) for the first episode of 2015. We discuss a number of interesting topics including the contributions Doubtful News makes to getting scientific and skeptical... by Data Skeptic
31 min listen
[MINI] Data Provenance: This episode introduces a high level discussion on the topic of Data Provenance, with more MINI episodes to follow to get into specific topics. Thanks to listener Sara L who wrote in to point out the Data Skeptic Podcast has focused alot... by Data Skeptic
11 min listen
[MINI] Selection Bias: A discussion about conducting US presidential election polls helps frame a converation about selection bias. by Data Skeptic
15 min listen
[MINI] z-scores: This week's episode dicusses z-scores, also known as standard score. This score describes the distance (in standard deviations) that an observation is away from the mean of the population. A closely related top is the 68-95-99.7 rule which... by Data Skeptic
10 min listen
Contest Announcement: The Data Skeptic Podcast is launching a contest- not one of chance, but one of skill. Listeners are encouraged to put their data science skills to good use, or if all else fails, guess! The contest works as follows. Below is some data about the... by Data Skeptic
12 min listen
The Secret and the Global Consciousness Project with Alex Boklin: I'm joined this week by Alex Boklin to explore the topic of magical thinking especially in the context of Rhonda Byrne's "The Secret", and the similarities it bears to The Global Consciousness Project (GCP). The GCP puts forward the hypothesis that... by Data Skeptic
42 min listen
Data Science at ZestFinance with Marick Sinay: Marick Sinay from ZestFianance is our guest this weel. This episode explores how data science techniques are applied in the financial world, specifically in assessing credit worthiness. by Data Skeptic
31 min listen

More Episodes from Data Skeptic

Skip carousel

Modelling Evolution: Modeling evolutionary processes goes way beyond the Hardy-Weinberg Equilibrium we all learned in biology class. Natural selection comes from many sources like resources availability, mate preferences, competition. Modeling entire populations of... by Data Skeptic
41 min listen
Behavioral Genetics: It’s almost impossible to think about animal behavior without thinking of dogs! Our canine friends are a subspecies of wolf that has been co-evolving with us for tens of thousands of years. The transition from wolf to pet has required intense... by Data Skeptic
47 min listen
Signal in the Noise: In this episode, we are joined by Barbara Webb and Anna Hadjitofi. Barbara runs the Insect Robotics lab at the University of Edinburgh, and Anna is a PhD student at the School of Informatics at the university. She is interested in studying and... by Data Skeptic
42 min listen
Pose Tracking: Many researchers and students have painstakingly labeled precise details about the body positions of the creatures they study. Can AI be used for this labeling? Of course it can! Today's episode discusses Social LEAP Estimates Animal Poses (SLEAP), a... by Data Skeptic
51 min listen
Modeling Group Behavior: Our guest in this episode is Sebastien Motsch, an assistant professor at Arizona State University, working in the School of Mathematical and Statistical Science. He works on modeling self-organized biological systems to understand how complex patterns... by Data Skeptic
41 min listen
Advances in Data Loggers: Our guest in this episode is Ryan Hanscom. Ryan is a Ph.D. candidate in a joint doctoral evolution program at San Diego State University and the University of California, Riverside. He is a terrestrial ecologist with a focus on herpetology and... by Data Skeptic
36 min listen
What You Know About Intelligence is Wrong (fixed): We are joined by Hank Schlinger, a professor of psychology at California State University, Los Angeles. His research revolves around theoretical issues in psychology and behavioral analysis. Hank establishes that words have references and... by Data Skeptic
42 min listen
Animal Decision Making: On today’s episode, we are joined by Aimee Dunlap. Aimee is an assistant professor at the University of Missouri–St. Louis and the interim director at the Whitney R. Harris World Ecology Center. Aimee discussed how animals perceive information and... by Data Skeptic
37 min listen
Octopus Cognition: We are joined by Tamar Gutnick, a visiting professor at the University of Naples Federico II, Napoli, Italy. She studies the octopus nervous system and their behavior, focusing on cognition and learning behaviors. Tamar gave a background to the kind... by Data Skeptic
38 min listen
Optimal Foraging: Claire Hemmingway, an assistant professor in the Department of Psychology and Ecology and Evolutionary Biology at the University of Tennessee in Knoxville, is our guest today. Her research is on decision-making in animal cognition, focusing on... by Data Skeptic
38 min listen
Memory in Chess: On today’s show, we are joined by our co-host, Becky Hansis-O’Neil. Becky is a Ph.D. student at the University of Missouri, St Louis, where she studies bumblebees and tarantulas to understand their learning and cognitive work. She joins us... by Data Skeptic
49 min listen
OpenWorm: On this episode, we are joined by Stephen Larson, the CEO of MetaCell and an affiliate of the OpenWorm foundation. Stephen discussed what the Openworm project is about. They hope to use a digital C. elegans nematode (C. elegans for... by Data Skeptic
34 min listen
What the Antlion Knows: Our guest is Becky Hansis-O’Neil, a Ph.D. student at the University of Missouri, St Louis, and our co-host for the new "Animal Intelligence" season. Becky shares her background on how she got into the field of behavioral intelligence and biology. by Data Skeptic
42 min listen
AI Roundtable: Kyle is joined by friends and former guests Pramit Choudhary and Frank Bell to have an open discussion of the impacts LLMs and machine learning have had in the past year on industry, and where things may go in the current year. by Data Skeptic
51 min listen
Uncontrollable AI Risks: We are joined by Darren McKee, a Policy Advisor and the host of Reality Check — a critical thinking podcast. Darren gave a background about himself and how he got into the AI space. Darren shared his thoughts on AGI's achievements in the coming... by Data Skeptic
39 min listen
I LLM and You Can Too: It took a massive financial investment for the first large language models (LLMs) to be created. Did their corporate backers lock these tools away for all but the richest? No. They provided comodity priced API options for using... by Data Skeptic
24 min listen
Q&A with Kyle: We celebrate episode 1000000000 with some Q&A from host Kyle Polich. We boil this episode down to four key questions: 1) How do you find guests 2) What is Data Skeptic all about? 3) What is Kyle all about? 4) What are Kyle's thoughts on AGI?... by Data Skeptic
40 min listen
LLMs for Data Analysis: In this episode, we are joined by Amir Netz, a Technical Fellow at Microsoft and the CTO of Microsoft Fabric. He discusses how companies can use Microsoft's latest tools for business intelligence. Amir started by discussing how business intelligence... by Data Skeptic
29 min listen
AI Platforms: Our guest today is Eric Boyd, the Corporate Vice President of AI at Microsoft. Eric joins us to share how organizations can leverage AI for faster development. Eric shared the benefits of using natural language to build products. He discussed the... by Data Skeptic
34 min listen
Deploying LLMs: We are excited to be joined by Aaron Reich and Priyanka Shah. Aaron is the CTO at Avanade, while Priyanka leads their AI/IoT offering for the SEA Region. Priyanka is also the MVP for Microsoft AI. They join us to discuss how LLMs are deployed in... by Data Skeptic
35 min listen
A Survey Assessing Github Copilot: In this episode, we are joined by Jenny Liang, a PhD student at Carnegie Mellon University, where she studies the usability of code generation tools. She discusses her recent survey on the usability of AI programming assistants. Jenny discussed the... by Data Skeptic
26 min listen
Program Aided Language Models: We are joined by Aman Madaan and Shuyan Zhou. They are both PhD students at the Language Technology Institute at Carnegie Mellon University. They join us to discuss their latest published paper, PAL: Program-aided Language Models. Aman and Shuyan... by Data Skeptic
32 min listen
Which Programming Language is ChatGPT Best At: In this episode, we have Alessio Buscemi, a software engineer at Lifeware SA. Alessio was a post-doctoral researcher at the University of Luxembourg. He joins us to discuss his paper, A Comparative Study of Code Generation using ChatGPT 3.5 across 10... by Data Skeptic
40 min listen
GraphText: On the show today, we are joined by Jianan Zhao, a Computer Science student at Mila and the University of Montreal. His research focus is on graph databases and natural language processing. He joins us to discuss how to use graphs with LLMs... by Data Skeptic
31 min listen
arXiv Publication Patterns: Today, we are joined by Rajiv Movva, a PhD student in Computer Science at Cornell Tech University. His research interest lies in the intersection of responsible AI and computational social science. He joins to discuss the findings of this work that... by Data Skeptic
28 min listen
Do LLMs Make Ethical Choices: We are excited to be joined by Josh Albrecht, the CTO of Imbue. Imbue is a research company whose mission is to create AI agents that are more robust, safer, and easier to use. He joins us to share findings of his work; Despite "super-human"... by Data Skeptic
29 min listen
Emergent Deception in LLMs: On today’s show, we are joined by Thilo Hagendorff, a Research Group Leader of Ethics of Generative AI at the University of Stuttgart. He joins us to discuss his research, . Thilo discussed how machine psychology is useful in machine learning... by Data Skeptic
27 min listen
Agents with Theory of Mind Play Hanabi: Nieves Montes, a Ph.D. student at the Artificial Intelligence Research Institute in Barcelona, Spain, joins us. Her PhD research revolves around value-based reasoning in relation to norms. She shares her latest study, Combining theory of mind and... by Data Skeptic
38 min listen
LLMs for Evil: We are joined by Maximilian Mozes, a PhD student at the University College, London. His PhD research focuses on Natural Language Processing (NLP), particularly the intersection of adversarial machine learning and NLP. He joins us to discuss his latest... by Data Skeptic
26 min listen

Related podcast episodes

Skip carousel

Film Conversations Episode Four: Ex Machina with Dr Isabel Millar: In today’s episode we speak with Dr. Isabel Millar, Author of The Psychoanalysis of Artificial Intelligence (published in 2021 - which I highly recommend). Dr. Millar is Associate Researcher at Newcastle University Department of Philosophy and both... by Film Conversations
81 min listen
Wallaby, damaged doubt, rat pilots: Julie-Anne Popple discusses the foraging behaviour of swamp wallabies with Miguel Bedoya-Perez. Ian Woolf finds out how damage to your brain can reduce your ability to doubt your beliefs. Julie-Anne Popple reports on dog shaking, cave spiders and denting by Diffusion Science radio
20 min listen
Dating science, brains and bacterial art: What are the seven secrets of successful online dating? Patrick Rubie and Ian Woolf unravel the mysteries 'Brain spindles that minds are made of' Nija Dalal talks to Charles Siebert about our brains. And Patrick Rubie reveals bacterial artists that draw by Diffusion Science radio
20 min listen
200 Tools, Blurry to Photorealistic, and Logical Neural Networks: This week we have our regular panel of Lan, George and Kyle! George brings us a blog post "What I Learned From Looking at 200 Machine Learning Tools." Which, the author, talks about a list complied of all of the AI/ML tools he could find. Lan brings... by Journal Club
33 min listen
MGG 543: You Learn Something New Every Day: It's been a crazy week, and the Apple event just adds to the mayhem. But no worries, your two favorite geeks are here to answer... by Mac Geek Gab — Your Questions Answered, Tips Shared, Troubleshooting Assistance
20 min listen
Episode 6 - Julian Togelius: In this episode of Eye on AI, I talk to Julian Togelius, perhaps the most prolific researcher at the intersection of video games and artificial intelligence. Julian works on AI for games and games for AI. Some of his most significant work is in... by Eye On A.I.
28 min listen
Chapter 13: Data Science Failures by Build a Career in Data Science
66 min listen
CM 153: Janelle Shane on How Artificial Intelligence Works: What happens when you teach an AI to write knock-knock jokes, recipes, and pick-up lines? - It's a rare week that goes by without someone talking about the power, and the perils, of artificial intelligence. by Curious Minds at Work
43 min listen
Miguel Nicolelis and Ronald Cicurel: The Singularity Isn’t Near and the Brain Can’t Be Simulated: The idea that man is a machine is an old one. And since the brain is a physical part of that machine it is often presumed that it can necessarily be simulated on a digital computer. But very little scientific scrutiny has been given on that presumption... by Singularity.FM
90 min listen
Brain spndle dating tips: From 2009: What are the seven secrets of successful online dating? Patrick Rubie and Ian Woolf unravel the mysteries; 'Brain spindles that minds are made of' Nija Dalal talks to Charles Siebert about our brains; And Patrick Rubie reveals bacterial artists that draw our attention in the latest science news Presented by Ian Woolf Co-produced by Patrick Rubie and Ian Woolf Support Diffusion by making a contribution Support Diffusion by buying Merchandise by Diffusion Science radio
20 min listen
Mind controlled brains, artificial IQ: Software beats humans on verbal IQ test by Ian Woolf, Laura Jade talks about her mind-controlled illuminated brain art, Sam Gentle explains how he helped make the mind-controlled colour-changing brain come to life. Production checked by Charles Willock, by Diffusion Science radio
20 min listen
Remote Design: In this week's episode, Kate and Laura yell about how to pretend you're still getting work done while you're actually searching online for toilet paper during a pandemic. Also, some tips about actually getting some design done from home, which is a... by What is Wrong with UX
32 min listen
Easily Fooling Deep Neural Networks: My guest this week is Anh Nguyen, a PhD student at the University of Wyoming working in the Evolving AI lab. The episode discusses the paper Deep Neural Networks are Easily Fooled [pdf] by Anh Nguyen, Jason Yosinski, and Jeff Clune. It... by Data Skeptic
28 min listen
CM 170: Nicholas Carr on What the Internet Does to Our Brains: Is the onslaught of online information eroding our brain's ability to think deeply and creatively? - In 2008, Nicholas Carr, asked the provocative question, "Is Google Making Us Stupid?" Two years later, he delved more deeply into this topic in his Pu... by Curious Minds at Work
51 min listen
How to Leave Your Dog at Home - Talking Dogs Radio Show: There are some ridiculous rituals that people go through with their dogs when they leave their house. But it doesn't have to be complicated! Takeaways: What normally happens that creates anxiety What should happen Why it needs to look this way ... by Vermont Dog Trainer Show
9 min listen
13. Hide your cats by Incubation Time
56 min listen
BI 122 Kohitij Kar: Visual Intelligence: Support the show to get full episodes and join the Discord community. Ko and I discuss a range of topics around his work to understand our visual intelligence. Ko was a postdoc in James Dicarlos lab, where he helped develop the convolutional neur by Brain Inspired
93 min listen
TWiN 49: Mirror, mirror on the wall, who's the fairest mouse of all?: TWiN welcomes mice to the elite club of ‘self-aware’ animals, with a study demonstrating a mirror-induced self-directed behavior in mice resembling visual self-recognition. Hosts: , , and Vivianne Morrison Subscribe (free):... by This Week in Neuroscience
91 min listen
Episode 61: Grab Bag 4: LAMOB Strikes Back: It's another solo-Jolie episode! While Dave is away, Jolie reaches into her grabbag of questions and pulls out queries about removing ink and tea stains from bedding, and what to do when your head discolors the wall behind your bed. To round out t... by Ask a Clean Person
47 min listen
Black Boxes Are Not Required: Deep neural networks are undeniably effective. They rely on such a high number of parameters, that they are appropriately described as “black boxes”. While black boxes lack desirably properties like interpretability and explainability, in some... by Data Skeptic
32 min listen
WikiLeaks and Vault 7: WikiLeaks and Vault 7 by CyberWire Daily
16 min listen
Trial by Pilot - Warrior Nun: We discuss the pilot to Warrior Nun, a new Netflix series based on the comics. We'll also take a thoughtful look at some deep philosophical questions. Can angels die? Why is Saturday CCD a thing? And why can't people properly light a Christmas tree? by Trial by Pilot
40 min listen
$17M Bitcoin blunder, Jeff Bezos' AI bet & lifesaving voicemail trick: Plus, it's your pet versus AI — scientists reveal who's really man's best friend. Hear how Natalie Nasatka's quick thinking with her Apple Watch turned life-saving. Also, Zillow's latest report is out, and it's the smaller towns, not big cities,... by Kim Komando Today
33 min listen
Nilly and the Duck by KooperKast
6 min listen
Ep. 517: Thinking Tools with Michael Covel on Trend Following Radio: One of Michael’s favorite things to do is intellectually kick people in the ass. People have stopped thinking. We have social media overload and virtual reality is taking over reality. Michael brings a clip from Daniel Dennett on the podcast.... by Michael Covel's Trend Following
23 min listen
Ep. 837: Nir Eyal Interview with Michael Covel on Trend Following Radio: Author Nir Eyal joins Michael for his second appearance on the show. And today it’s all about distraction. For example. You sit down at your desk to work on an important project, but a notification on your phone interrupts your morning. Later, as... by Michael Covel's Trend Following
51 min listen
Brain Dust: ENCORE Know your brain? Think again. Driven by a hidden agenda, powered by an indecipherable web of neurons, and influenced by other brains, your grey matter is a black box. To "know thyself" may be a challenge, and free will... by Big Picture Science
54 min listen
China: This week we’re breaking down “China.” When Michael outsmarts Oscar in front of the office, Oscar can’t let it go. Meanwhile Andy keeps bothering Darryl via text messages. Jenna goes over texting etiquette, Angela does some real Mom Detectives’ work on finding the lunch spot that’s featured on Dwight’s mug and we learn how Phil Shea made that toilet paper spooling machine. We also have a little treat from The Office’s first assistant director, Kelly Cantley. So BTB and enjoy this episode! by Office Ladies
71 min listen
144: Mobile Kids, Mom Madness, and Outgrowing The Bad: Kail and Lindsie are back and starting off strong with another wild animal story from Kail. What is with her house and attracting animals? Then, and babies start to move around more they get harder to handle. Kail and Lindsie talk about,... by Coffee Convos with Kail Lowry and Lindsie Chrisley
70 min listen
Scott in Plastic, He's Fantastic!: Dolls by Keep It Weird
109 min listen

Discover this podcast and so much more

Computer Vision is Not Perfect

Computer Vision is Not Perfect

Description

Titles in the series (100)

More Episodes from Data Skeptic

Related podcast episodes