The Data Skeptic Podcast features interviews and discussion of 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 veracity of claims and efficacy of approaches.
Walt Woods joins us to discuss his paper Adversarial Explanations for Understanding Image Classification Decisions and Improved Neural Network Robustness with co-authors Jack Chen and Christof Teuscher.
Andrei Barbu joins us to discuss ObjectNet - a new kind of vision dataset. In contrast to ImageNet, ObjectNet seeks to provide images that are more representative of the types of images an autonomous machine is likely to encounter in the real world. Collecting a dataset in this way required careful use of Mechanical Turk to get Turkers to provide a corpus of images that removes some of the bias found in ImageNet. http://0xab.com/
Enrico Bertini joins us to discuss how data visualization can be used to help make machine learning more interpretable and explainable. Find out more about Enrico at http://enrico.bertini.io/. More from Enrico with co-host Moritz Stefaner on the Data Stories podcast!
Wiebe van Ranst joins us to talk about a project in which specially designed printed images can fool a computer vision system, preventing it from identifying a person. Their attack targets the popular YOLO2 pre-trained image recognition model, and thus, is likely to be widely applicable.
Interpretability Machine learning has shown a rapid expansion into every sector and industry. With increasing reliance on models and increasing stakes for the decisions of models, questions of how models actually work are becoming increasingly important to ask. Welcome to Data Skeptic Interpretability. In this episode, Kyle interviews Christoph Molnar about his book Interpretable Machine Learning. Thanks to our sponsor, the Gartner Data & Analytics Summit going on in Grapevine, TX on March 23 – 26, 2020. Use discount code: dataskeptic. Music Our new theme song is #5 by Big D and the Kids Table. Incidental music by Tanuki Suit Riot.
Alex Reeves joins us to discuss some of the challenges around building a serverless, scalable, generic machine learning pipeline. The is a technical deep dive on architecting solutions and a discussion of some of the design choices made.
The modern deep learning approaches to natural language processing are voracious in their demands for large corpora to train on. Folk wisdom estimates used to be around 100k documents were required for effective training. The availability of broadly trained, general-purpose models like BERT has made it possible to do transfer learning to achieve novel results on much smaller corpora. Thanks to these advancements, an NLP researcher might get value out of fewer examples since they can use the transfer learning to get a head start and focus on learning the nuances of the language specifically relevant to the task at hand. Thus, small specialized corpora are both useful and practical to create. In this episode, Kyle speaks with Mor Geva, lead author on the recent paper Are We Modeling the Task or the Annotator? An Investigation of Annotator Bias in Natural Language Understanding Datasets, which explores some unintended consequences of the typical procedure followed for generating corpora. Source code for the paper available here: https://github.com/mega002/annotator_bias
While at MS Build 2019, Kyle sat down with Lance Olson from the Applied AI team about how tools like cognitive services and cognitive search enable non-data scientists to access relatively advanced NLP tools out of box, and how more advanced data scientists can focus more time on the bigger picture problems.
Manuel Mager joins us to discuss natural language processing for low and under-resourced languages. We discuss current work in this area and the Naki Project which aggregates research on NLP for native and indigenous languages of the American continent.
GPT-2 is yet another in a succession of models like ELMo and BERT which adopt a similar deep learning architecture and train an unsupervised model on a massive text corpus. As we have been covering recently, these approaches are showing tremendous promise, but how close are they to an AGI? Our guest today, Vazgen Davidyants wondered exactly that, and have conversations with a Chatbot running GPT-2. We discuss his experiences as well as some novel thoughts on artificial intelligence.
Tim Niven joins us this week to discuss his work exploring the limits of what BERT can do on certain natural language tasks such as adversarial attacks, compositional learning, and systematic learning.
Video annotation is an expensive and time-consuming process. As a consequence, the available video datasets are useful but small. The availability of machine transcribed explainer videos offers a unique opportunity to rapidly develop a useful, if dirty, corpus of videos that are "self annotating", as hosts explain the actions they are taking on the screen. This episode is a discussion of the HowTo100m dataset - a project which has assembled a video corpus of 136M video clips with captions covering 23k activities. Related Links The paper will be presented at ICCV 2019 @antoine77340 Antoine on Github Antoine's homepage
In 2017, Facebook published a paper called Deal or No Deal? End-to-End Learning for Negotiation Dialogues. In this research, the reinforcement learning agents developed a mechanism of communication (which could be called a language) that made them able to optimize their scores in the negotiation game. Many media sources reported this as if it were a first step towards Skynet taking over. In this episode, Kyle discusses bargaining agents and the actual results of this research.
Priyanka Biswas joins us in this episode to discuss natural language processing for languages that do not have as many resources as those that are more commonly studied such as English. Successful NLP projects benefit from the availability of like large corpora, well-annotated corpora, software libraries, and pre-trained models. For languages that researchers have not paid as much attention to, these tools are not always available.
Kyle and Linh Da discuss the class of approaches called "Named Entity Recognition" or NER. NER algorithms take any string as input and return a list of "entities" - specific facts and agents in the text along with a classification of the type (e.g. person, date, place).
USC students from the CAIS++ student organization have created a variety of novel projects under the mission statement of "artificial intelligence for social good". In this episode, Kyle interviews Zane and Leena about the Endangered Languages Project.
When users on Twitter post with geographic tags, it creates the opportunity for a variety of interesting questions to be posed having to do with language, dialects, and location. In this episode, Kyle interviews Bruno Gonçalves about his work studying language in this way.
Modern messaging technology has facilitated a trend towards highly compact, short messages send by users who can presume a great amount of context held between the communicating parties. The rules of grammar may be discarded and often visible errors are a normal part of the conversation. >>> Good mornink >>> morning Yet such short messages are also important for businesses whose users are unlikely to read a large block of text upon completing an order. Similarly, a business might want to offer assistance and effective question and answering solutions in an automated and ideally multi-lingual way. In this episode, we discuss techniques for designing solutions like that.
ELMo (Embeddings from Language Models) introduced the idea of deep contextualized word representations. It extends previous ideas like word2vec and GloVe. The ELMo model is a neural network able to map natural language into a vector space. This vector space, out of box, proved to be incredibly useful in a wide variety of seemingly unrelated NLP tasks like sentiment analysis and name entity recognition.
Bilingual evaluation understudy (or BLEU) is a metric for evaluating the quality of machine translation using human translation as examples of acceptable quality results. This metric has become a widely used standard in the research literature. But is it the perfect measure of quality of machine translation?
Machine transcription (the process of translating audio recordings of language to text) has come a long way in recent years. But how do the errors made during machine transcription compare to the errors made by a human transcriber? Find out in this episode!
A sequence to sequence (or seq2seq) model is neural architecture used for translation (and other tasks) which consists of an encoder and a decoder. The encoder/decoder architecture has obvious promise for machine translation, and has been successfully applied this way. Encoding an input to a small number of hidden nodes which can effectively be decoded to a matching string requires machine learning to learn an efficient representation of the essence of the strings. In addition to translation, seq2seq models have been used in a number of other NLP tasks such as summarization and image captioning. Related Links tf-seq2seq Describing Multimedia Content using Attention-based Encoder--Decoder Networks Show and Tell: A Neural Image Caption Generator Attend to You: Personalized Image Captioning with Context Sequence Memory Networks
Kyle interviews Julia Silge about her path into data science, her book Text Mining with R, and some of the ways in which she's used natural language processing in projects both personal and professional. Related Links https://stack-survey-2018.glitch.me/ https://stackoverflow.blog/2017/03/28/realistic-developer-fiction/
One of the most challenging NLP tasks is natural language understanding and reasoning. How can we construct algorithms that are able to achieve human level understanding of text and be able to answer general questions about it? This is truly an open problem, and one with the bAbI dataset has been constructed to facilitate. bAbI presents a variety of different language understanding and reasoning tasks and exists as benchmark for comparing approaches. In this episode, Kyle talks to Rasmus Berg Palm about his recent paper Recurrent Relational Networks
In the first half of this episode, Kyle speaks with Marc-Alexandre Côté and Wendy Tay about Text World. Text World is an engine that simulates text adventure games. Developers are encouraged to try out their reinforcement learning skills building agents that can programmatically interact with the generated text adventure games. In the second half of this episode, Kyle interviews Kevin Patel about his paper Towards Lower Bounds on Number of Dimensions for Word Embeddings. In this research, the explore an important question of how many hidden nodes to use when creating a word embedding.
Word2vec is an unsupervised machine learning model which is able to capture semantic information from the text it is trained on. The model is based on neural networks. Several large organizations like Google and Facebook have trained word embeddings (the result of word2vec) on large corpora and shared them for others to use. The key algorithmic ideas involved in word2vec is the continuous bag of words model (CBOW). In this episode, Kyle uses excerpts from the 1983 cinematic masterpiece War Games, and challenges Linhda to guess a word Kyle leaves out of the transcript. This is similar to how word2vec is trained. It trains a neural network to predict a hidden word based on the words that appear before and after the missing location.
In a recent paper, Leveraging Discourse Information Effectively for Authorship Attribution, authors Su Wang, Elisa Ferracane, and Raymond J. Mooney describe a deep learning methodology for predict which of a collection of authors was the author of a given document.
The earliest efforts to apply machine learning to natural language tended to convert every token (every word, more or less) into a unique feature. While techniques like stemming may have cut the number of unique tokens down, researchers always had to face a problem that was highly dimensional. Naive Bayes algorithm was celebrated in NLP applications because of its ability to efficiently process highly dimensional data. Of course, other algorithms were applied to natural language tasks as well. While different algorithms had different strengths and weaknesses to different NLP problems, an early paper titled Scaling to Very Very Large Corpora for Natural Language Disambiguation popularized one somewhat surprising idea. For many NLP tasks, simply providing a large corpus of examples not only improved accuracy, but it also showed that asymptotically, some algorithms yielded more improvement from working on very, very large corpora. Although not explicitly in about NLP, the noteworthy paper The Unreasonable Effectiveness of Data emphasizes this point further while paying homage to the classic treatise The Unreasonable Effectiveness of Mathematics in the Natural Sciences. In this episode, Kyle shares a few thoughts along these lines with Linh Da. The discussion winds up with a brief introduction to Zipf's law. When applied to natural language, Zipf's law states that the frequency of any given word in a corpus (regardless of language) will be proportional to its rank in the frequency table.
Github is many things besides source control. It's a social network, even though not everyone realizes it. It's a vast repository of code. It's a ticketing and project management system. And of course, it has search as well. In this episode, Kyle interviews Hamel Husain about his research into semantic code search.
This episode reboots our podcast with the theme of Natural Language Processing for the next few months. We begin with introductions of Yoshi and Linh Da and then get into a broad discussion about natural language processing: what it is, what some of the classic problems are, and just a bit on approaches. Finishing out the show is an interview with Lucy Park about her work on the KoNLPy library for Korean NLP in Python. If you want to share your NLP project, please join our Slack channel. We're eager to see what listeners are working on! http://konlpy.org/en/latest/
In today's episode, Kyle chats with Alexander Zhebrak, CTO of Insilico Medicine, Inc. Insilico self describes as artificial intelligence for drug discovery, biomarker development, and aging research. The conversation in this episode explores the ways in which machine learning, in particular, deep learning, is contributing to the advancement of drug discovery. This happens not just through research but also through software development. Insilico works on data pipelines and tools like MOSES, a benchmarking platform to support research on machine learning for drug discovery. The MOSES platform provides a standardized benchmarking dataset, a set of open-sourced models with unified implementation, and metrics to evaluate and assess their performance.
At the NeurIPS 2018 conference, Stradigi AI premiered a training game which helps players learn American Sign Language. This episode brings the first of many interviews conducted at NeurIPS 2018. In this episode, Kyle interviews Chief Data Scientist Carolina Bessega about the deep learning architecture used in this project. The Stradigi AI team was exhibiting a project called the American Sign Language (ASL) Alphabet Game at the recent NeurIPS 2018 conference. They also published a detailed blog post about how they built the system found here.
This week, Kyle interviews Scott Nestler on the topic of Data Ethics. Today, no ubiquitous, formal ethical protocol exists for data science, although some have been proposed. One example is the INFORMS Ethics Guidelines. Guidelines like this are rather informal compared to other professions, like the Hippocratic Oath. Yet not every profession requires such a formal commitment. In this episode, Scott shares his perspective on a variety of ethical questions specific to data and analytics.
Kyle interviews Mick West, author of Escaping the Rabbit Hole: How to Debunk Conspiracy Theories Using Facts, Logic, and Respect about the nature of conspiracy theories, the people that believe them, and how to help people escape the belief in false information. Mick is also the creator of metabunk.org. The discussion explores conspiracies like chemtrails, 9/11 conspiracy theories, JFK assassination theories, and the flat Earth theory. We live in a complex world in which no person can have a sufficient understanding of all topics. It's only natural that some percentage of people will eventually adopt fringe beliefs. In this book, Mick provides a fantastic guide to helping individuals who have fallen into a rabbit hole of pseudo-science or fake news.
Fake news attempts to lead readers/listeners/viewers to conclusions that are not descriptions of reality. They do this most often by presenting false premises, but sometimes by presenting flawed logic. An argument is only sound and valid if the conclusions are drawn directly from all the state premises, and if there exists a path of logical reasoning leading from those premises to the conclusion. While creating a theorem does feel to most mathematicians as a creative act of discovery, some theorems have been proven using nothing more than search. All the "rules" of logic (like modus ponens) can be encoded into a computer program. That program can start from the premises, applying various combinations of rules to inference new information, and check to see if the program has inference the desired conclusion or its negation. This does seem like a mechanical process when painted in this light. However, several challenges exist preventing any theorem prover from instantly solving all the open problems in mathematics. In this episode, we discuss a bit about what those challenges are.
Fake news can be responded to with fact-checking. However, it's easier to create fake news than the fact check it. Full Fact is the UK's independent fact-checking organization. In this episode, Kyle interviews Mevan Babakar, head of automated fact-checking at Full Fact. Our discussion talks about the process and challenges in doing fact-checking. Full Fact has been exploring ways in which machine learning can assist in automating parts of the fact-checking process. Progress in areas like this allows journalists to be more effective and rapid in responding to new information.
In mathematics, truth is universal. In data, truth lies in the where clause of the query. As large organizations have grown to rely on their data more significantly for decision making, a common problem is not being able to agree on what the data is. As the volume and velocity of data grow, challenges emerge in answering questions with precision. A simple question like "what was the revenue yesterday" could become mired in details. Did your query account for transactions that haven't been finalized? If I query again later, should I exclude orders that have been returned since the last query? What time zone should I use? The list goes on and on. In any large enough organization, you are also likely to find multiple copies if the same data. Independent systems might record the same information with slight variance. Sometimes systems will import data from other systems; a process which could become out of sync for several reasons. For any sufficiently large system, answering analytical questions with precision can become a non-trivial challenge. The business intelligence community aspires to provide a "single source of truth" - one canonical place where data consumers can go to get precise, reliable, and trusted answers to their analytical questions.
Fast radio bursts are an astrophysical phenomenon first observed in 2007. While many observations have been made, science has yet to explain the mechanism for these events. This has led some to ask: could it be a form of extra-terrestrial communication? Probably not. Kyle asks Gerry Zhang who works at the Berkeley SETI Research Center about this possibility and more importantly, about his applications of deep learning to detect fast radio bursts. Radio astronomy captures observations from space which can be converted to a waterfall chart or spectrogram. These data structures can be formatted in a visual way and also make great candidates for applying deep learning to the task of detecting the fast radio bursts.
This episode explores the root concept of what it is to be Bayesian: describing knowledge of a system probabilistically, having an appropriate prior probability, know how to weigh new evidence, and following Bayes's rule to compute the revised distribution. We present this concept in a few different contexts but primarily focus on how our bird Yoshi sends signals about her food preferences. Like many animals, Yoshi is a complex creature whose preferences cannot easily be summarized by a straightforward utility function the way they might in a textbook reinforcement learning problem. Her preferences are sequential, conditional, and evolving. We may not always know what our bird is thinking, but we have some good indicators that give us clues.
This is our interview with Dorje Brody about his recent paper with David Meier, How to model fake news. This paper uses the tools of communication theory and a sub-topic called filtering theory to describe the mathematical basis for an information channel which can contain fake news. Thanks to our sponsor Gartner.
Without getting into definitions, we have an intuitive sense of what a "community" is. The Louvain Method for Community Detection is one of the best known mathematical techniques designed to detect communities. This method requires typical graph data in which people are nodes and edges are their connections. It's easy to imagine this data in the context of Facebook or LinkedIn but the technique applies just as well to any other dataset like cellular phone calling records or pen-pals. The Louvain Method provides a means of measuring the strength of any proposed community based on a concept known as Modularity. Modularity is a value in the range that measure the density of links internal to a community against links external to the community. The quite palatable assumption here is that a genuine community would have members that are strongly interconnected. A community is not necessarily the same thing as a clique; it is not required that all community members know each other. Rather, we simply define a community as a graph structure where the nodes are more connected to each other than connected to people outside the community. It's only natural that any person in a community has many connections to people outside that community. The more a community has internal connections over external connections, the stronger that community is considered to be. The Louvain Method elegantly captures this intuitively desirable quality.
In this episode, our guest is Dan Kahan about his research into how people consume and interpret science news. In an era of fake news, motivated reasoning, and alternative facts, important questions need to be asked about how people understand new information. Dan is a member of the Cultural Cognition Project at Yale University, a group of scholars interested in studying how cultural values shape public risk perceptions and related policy beliefs. In a paper titled Cultural cognition of scientific consensus, Dan and co-authors Hank Jenkins‐Smith and Donald Braman discuss the "cultural cognition of risk" and establish experimentally that individuals tend to update their beliefs about scientific information through a context of their pre-existing cultural beliefs. In this way, topics such as climate change, nuclear power, and conceal-carry handgun permits often result in people. The findings of this and other studies tell us that on topics such as these, even when people are given proper information about a scientific consensus, individuals still interpret those results through the lens of their pre-existing cultural beliefs. The ‘cultural cognition of risk’ refers to the tendency of individuals to form risk perceptions that are congenial to their values. The study presents both correlational and experimental evidence confirming that cultural cognition shapes individuals’ beliefs about the existence of scientific consensus, and the process by which they form such beliefs, relating to climate change, the disposal of nuclear wastes, and the effect of permitting concealed possession of handguns. The implications of this dynamic for science communication and public policy‐making are discussed.
A false discovery rate (FDR) is a methodology that can be useful when struggling with the problem of multiple comparisons. In any experiment, if the experimenter checks more than one dependent variable, then they are making multiple comparisons. Naturally, if you make enough comparisons, you will eventually find some correlation. Classically, people applied the Bonferroni Correction. In essence, this procedure dictates that you should lower your p-value (raise your standard of evidence) by a specific amount depending on the number of variables you're considering. While effective, this methodology is strict about preventing false positives (type i errors). You aren't likely to find evidence for a hypothesis that is actually false using Bonferroni. However, your exuberance to avoid type i errors may have introduced some type ii errors. There could be some hypotheses that are actually true, which you did not notice. This episode covers an alternative known as false discovery rates. The essence of this method is to make more specific adjustments to your expectation of what p-value is sufficient evidence.
Digital videos can be described as sequences of still images and associated audio. Audio is easy to fake. What about video? A video can easily be broken down into a sequence of still images replayed rapidly in sequence. In this context, videos are simply very high dimensional sequences of observations, ripe for input into a machine learning algorithm. The availability of commodity hardware, clever algorithms, and well-designed software to implement those algorithms at scale make it possible to do machine learning on video, but to what end? There are many answers, one interesting approach being the technology called "DeepFakes". The Deep of Deepfakes refers to Deep Learning, and the fake refers to the function of the software - to take a real video of a human being and digitally alter their face to match someone else's face. Here are two examples: Barack Obama via Jordan Peele The versatility of Nick Cage This software produces curiously convincing fake videos. Yet, there's something slightly off about them. Surely machine learning can be used to determine real from fake... right? Siwei Lyu and his collaborators certainly thought so and demonstrated this idea by identifying a novel, detectable feature which was commonly missing from videos produced by the Deep Fakes software. In this episode, we discuss this use case for deep learning, detecting fake videos, and the threat of fake videos in the future.
Two weeks ago we discussed click through rates or CTRs and their usefulness and limits as a metric. Today, we discuss a related metric known as quality score. While that phrase has probably been used to mean dozens of different things in different contexts, our discussion focuses around the idea of quality score encountered in Search Engine Marketing (SEM). SEM is the practice of purchasing keyword targeted ads shown to customers using a search engine. Most SEM is managed via an auction mechanism - the advertiser states the price they are willing to pay, and in real time, the search engine will serve users advertisements and charge the advertiser. But how to search engines decide who to show and what price to charge? This is a complicated question requiring a multi-part answer to address completely. In this episode, we focus on one part of that equation, which is the quality score the search engine assigns to the ad in context. This quality score is calculated via several factors including crawling the destination page (also called the landing page) and predicting how applicable the content found there is to the ad itself.
Kyle interviews Steven Sloman, Professor in the school of Cognitive, Linguistic, and Psychological Sciences at Brown University. Steven is co-author of The Knowledge Illusion: Why We Never Think Alone and Causal Models: How People Think about the World and Its Alternatives. Steven shares his perspective and research into how people process information and what this teaches us about the existence of and belief in fake news.
A Click Through Rate (CTR) is the proportion of clicks to impressions of some item of content shared online. This terminology is most commonly used in digital advertising but applies just as well to content websites might choose to feature on their homepage or in search results. A CTR is intuitively appealing as a metric for optimization. After all, if users are disinterested in some content, under normal circumstances, it's reasonable to assume they would ignore the content, rather than clicking on it. On the other hand, the best content is likely to elicit a high CTR as users signal their interest by following the hyperlink. In the advertising world, a website could charge per impression, per click, or per action. Both impression and action based pricing have asymmetrical results for the publisher and advertiser. However, paying per click (CPC based advertising) seems to strike a nice balance. For this and other numeric reasons, many digital advertising mechanisms (such as Google Adwords) use CPC as the payment mechanism. When charging per click, an advertising platform will value a high CTR when selecting which ad to show. As we learned in our episode on Goodhart's Law, once a measure is turned into a target, it ceases to be a good measure. While CTR alone does not entirely drive most online advertising algorithms, it does play an important role. Thus, advertisers are incentivized to adopt strategies that maximize CTR. On the surface, this sounds like a great idea: provide internet users what they are looking for, and be awarded with their attention and lower advertising costs. However, one possible unintended consequence of this type of optimization is the creation of ads which are designed solely to generate clicks, regardless of if the users are happy with the page they visit after clicking a link. So, at least in part, websites that optimize for higher CTRs are going to favor content that does a good job getting viewers to click it. Getting a user to view a page is not totally synonymous with getting a user to appreciate the content of a page. The gap between the algorithmic goal and the user experience could be one of the factors that has promoted the creation of fake news.
The scale and frequency with which information can be distributed on social media makes the problem of fake news a rapidly metastasizing issue. To do any content filtering or labeling demands an algorithmic solution. In today's episode, Kyle interviews Kai Shu and Mike Tamir about their independent work exploring the use of machine learning to detect fake news. Kai Shu and his co-authors published Fake News Detection on Social Media: A Data Mining Perspective, a research paper which both surveys the existing literature and organizes the structure of the problem in a robust way. Mike Tamir led the development of fakerfact.org, a website and Chrome/Firefox plugin which leverages machine learning to try and predict the category of a previously unseen web page, with categories like opinion, wiki, and fake news.
If you prepared a list of creatures regarded as highly intelligent, it's unlikely ants would make the cut. This is expected, as on an individual level, ants do not generally display behavior that most humans would regard as intelligence. In fact, it might even be true that most species of ants are unable to learn. Despite this, ant colonies have evolved excellent survival mechanisms through the careful orchestration of ants.
With publications such as "Prior exposure increases perceived accuracy of fake news", "Lazy, not biased: Susceptibility to partisan fake news is better explained by lack of reasoning than by motivated reasoning", and "The science of fake news", Gordon Pennycook is asking and answering analytical questions about the nature of human intuition and fake news. Gordon appeared on Data Skeptic in 2016 to discuss people's ability to recognize pseudo-profound bullshit. This episode explores his work in fake news.
Today's spam filters are advanced data driven tools. They rely on a variety of techniques to effectively and often seamlessly filter out junk email from good email. Whitelists, blacklists, traffic analysis, network analysis, and a variety of other tools are probably employed by most major players in this area. Naturally content analysis can be an especially powerful tool for detecting spam. Given the binary nature of the problem ( or ) its clear that this is a great problem to use machine learning to solve. In order to apply machine learning, you first need a labelled training set. Thankfully, many standard corpora of labelled spam data are readily available. Further, if you're working for a company with a spam filtering problem, often asking users to self-moderate or flag things as spam can be an effective way to generate a large amount of labels for "free". With a labeled dataset in hand, a data scientist working on spam filtering must next do feature engineering. This should be done with consideration of the algorithm that will be used. The Naive Bayesian Classifer has been a popular choice for detecting spam because it tends to perform pretty well on high dimensional data, unlike a lot of other ML algorithms. It also is very efficient to compute, making it possible to train a per-user Classifier if one wished to. While we might do some basic NLP tricks, for the most part, we can turn each word in a document (or perhaps each bigram or n-gram in a document) into a feature. The Naive part of the Naive Bayesian Classifier stems from the naive assumption that all features in one's analysis are considered to be independent. If and are known to be independent, then . In other words, you just multiply the probabilities together. Shh, don't tell anyone, but this assumption is actually wrong! Certainly, if a document contains the word algorithm, it's more likely to contain the word probability than some randomly selected document. Thus, , violating the assumption. Despite this "flaw", the Naive Bayesian Classifier works remarkably will on many problems. If one employs the common approach of converting a document into bigrams (pairs of words instead of single words), then you can capture a good deal of this correlation indirectly. In the final leg of the discussion, we explore the question of whether or not a Naive Bayesian Classifier would be a good choice for detecting fake news.
How does fake news get spread online? Its not just a matter of manipulating search algorithms. The social platforms for sharing play a major role in the distribution of fake news. But how significant of an impact can there be? How significantly can bots influence the spread of fake news? In this episode, Kyle interviews Filippo Menczer, Professor of Computer Science and Informatics. Fil is part of the Observatory on Social Media ([OSoMe][https://osome.iuni.iu.edu/tools/]). OSoMe are the creators of Hoaxy, Botometer, Fakey, and other tools for studying the spread of information on social media. The interview explores these tools and the contributions Bots make to the spread of fake news.
This episode kicks off our new theme of "Fake News" with guests Robert Sheaffer and Brad Schwartz. Fake news is a new label for an old idea. For our purposes, we will define fake news information created to deliberately mislead while masquerading as a legitimate, journalistic source of truth. It's become a modern topic of discussion as our cultures evolve to the fledgling mechanisms of communication introduced by online platforms. What was the earliest incident of fake news? That's a question for which we may never find a satisfying answer. While not the earliest, we present a dramatization of an early example of fake news, which leads us into a discussion with UFO Skeptic Robert Sheaffer. Following that we get into our main interview with Brad Schwartz, author of Broadcast Hysteria: Orson Welles's War of the Worlds and the Art of Fake News.
We revisit the 2018 Microsoft Build in this episode, focusing on the latest ideas in DevOps. Kyle interviews Cloud Developer Advocates Damien Brady, Paige Bailey, and Donovan Brown to talk about DevOps and data science and databases. For a data scientist, what does it even mean to “build”? Packaging and deployment are things that a data scientist doesn't normally have to consider in their day-to-day work. The process of making an AI app is usually divided into two streams of work: data scientists building machine learning models and app developers building the application for end users to consume. DevOps includes all the parties involved in getting the application deployed and maintained and thinking about all the phases that follow and precede their part of the end solution. So what does DevOps mean for data science? Why should you adopt DevOps best practices? In the first half, Paige and Damian share their views on what DevOps for data science would look like and how it can be introduced to provide continuous integration, delivery, and deployment of data science models. In the second half, Donovan and Damian talk about the DevOps life cycle of putting a database under version control and carrying out deployments through a release pipeline.
Logic is a fundamental of mathematical systems. It's roots are the values true and false and it's power is in what it's rules allow you to prove. Prepositional logic provides it's user variables. This episode gets into First Order Logic, an extension to prepositional logic.
An intelligent agent trained in a simulated environment may be prone to making mistakes in the real world due to discrepancies between the training and real-world conditions. The areas where an agent makes mistakes are hard to find, known as "blind spots," and can stem from various reasons. In this week’s episode, Kyle is joined by Ramya Ramakrishnan, a PhD candidate at MIT, to discuss the idea “blind spots” in reinforcement learning and approaches to discover them.
In this week’s episode, our host Kyle interviews Gokula Krishnan from ETH Zurich, about his recent contributions to defenses against adversarial attacks. The discussion centers around his latest paper, titled “Defending Against Adversarial Attacks by Leveraging an Entire GAN,” and his proposed algorithm, aptly named ‘Cowboy.’
On a long car ride, Linhda and Kyle record a short episode. This discussion is about transfer learning, a technique using in machine learning to leverage training from one domain to have a head start learning in another domain. Transfer learning has some obvious appealing features. Take the example of an image recognition problem. There are now many widely available models that do general image recognition. Detecting that an image contains a "sofa" is an impressive feat. However, for a furniture company interested in more specific details, this classifier is absurdly general. Should the furniture company build a massive corpus of tagged photos, effectively starting from scratch? Or is there a way they can transfer the learnings from the general task to the specific one. A general definition of transfer learning in machine learning is the use of taking some or all aspects of a pre-trained model as the basis to begin training a new model which a specific and potentially limited dataset.
Medical imaging is a highly effective tool used by clinicians to diagnose a wide array of diseases and injuries. However, it often requires exceptionally trained specialists such as radiologists to interpret accurately. In this episode of Data Skeptic, our host Kyle Polich is joined by Gabriel Maicas, a PhD candidate at the University of Adelaide, to discuss machine learning systems that can be used by radiologists to improve their accuracy and speed of diagnosis.
Thanks to our sponsor Galvanize A Kalman Filter is a technique for taking a sequence of observations about an object or variable and determining the most likely current state of that object. In this episode, we discuss it in the context of tracking our lilac crowned amazon parrot Yoshi. Kalman filters have many applications but the one of particular interest under our current theme of artificial intelligence is to efficiently update one's beliefs in light of new information. The Kalman filter is based upon the Gaussian distribution. This distribution is described by two parameters: (the mean) and standard deviation. The procedure for updating these values in light of new information has a closed form. This means that it can be described with straightforward formulae and computed very efficiently. You may gain a greater appreciation for Kalman filters by considering what would happen if you could not rely on the Gaussian distribution to describe your posterior beliefs. If determining the probability distribution over the variables describing some object cannot be efficiently computed, then by definition, maintaining the most up to date posterior beliefs can be a significant challenge. Kyle will be giving a talk at Skeptical 2018 in Berkeley, CA on June 10.
There's so much to discuss on the AI side, it's hard to know where to begin. Luckily, Steve Guggenheimer, Microsoft’s corporate vice president of AI Business, and Carlos Pessoa, a software engineering manager for the company’s Cloud AI Platform, talked to Kyle about announcements related to AI in industry.
Thanks to our sponsor The Great Courses. This week's episode is a short primer on game theory. For tickets to the free Data Skeptic meetup in Chicago on Tuesday, May 15 at the Mendoza College of Business (224 South Michigan Avenue, Suite 350), click here,
In this episode of Data Skeptic, Kyle chats with Jerry Schwarz from the Independent Investigations Group (IIG)'s SF Bay Area chapter about testing claims of the paranormal. The IIG is a volunteer-based organization dedicated to investigating paranormal or extraordinary claim from a scientific viewpoint. The group, headquartered at the Center for Inquiry-Los Angeles in Hollywood, offers a $100,000 prize to anyone who can show, under proper observing conditions, evidence of any paranormal, supernatural, or occult power or event. CHICAGO Tues, May 15, 6pm. Come to our Data Skeptic meetup. CHICAGO Saturday, May 19, 10am. Kyle will be giving a talk at the Chicago AI, Data Science, and Blockchain Conference 2018.
Our guest this week, Hector Levesque, joins us to discuss an alternative way to measure a machine’s intelligence, called Winograd Schemas Challenge. The challenge was proposed as a possible alternative to the Turing test during the 2011 AAAI Spring Symposium. The challenge involves a small reading comprehension test about common sense knowledge.
This week on Data Skeptic, we begin with a skit to introduce the topic of this show: The Imitation Game. We open with a scene in the distant future. The year is 2027, and a company called Shamony is announcing their new product, Ada, the most advanced artificial intelligence agent. To prove its superiority, the lead scientist announces that it will use the Turing Test that Alan Turing proposed in 1950. During this we introduce Turing’s “objections” outlined in his famous paper, “Computing Machinery and Intelligence.” Following that, we talk with improv coach Holly Laurent on the art of improvisation and Peter Clark from the Allen Institute for Artificial Intelligence about question and answering algorithms.