Data Skeptic (general)

Derek Lim joins us to discuss the paper Expertise and Dynamics within Crowdsourced Musical Knowledge Curation: A Case Study of the Genius Platform.

 

Direct download: crowdsourced-expertise.mp3
Category:general -- posted at: 7:00am PDT

Neil Johnson joins us to discuss the paper The online competition between pro- and anti-vaccination views.

Direct download: the-spread-of-misinformation-online.mp3
Category:general -- posted at: 7:00am PDT



Direct download: consensus-voting.mp3
Category:general -- posted at: 7:00am PDT

Steven Heilman joins us to discuss his paper Designing Stable Elections.

For a general interest article, see: https://theconversation.com/the-electoral-college-is-surprisingly-vulnerable-to-popular-vote-changes-141104

Steven Heilman receives funding from the National Science Foundation. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation.

Direct download: voting-mechanisms.mp3
Category:general -- posted at: 7:00am PDT

Sami Yousif joins us to discuss the paper The Illusion of Consensus: A Failure to Distinguish Between True and False Consensus. This work empirically explores how individuals evaluate consensus under different experimental conditions reviewing online news articles.

More from Sami at samiyousif.org

Link to survey mentioned by Daniel Kerrigan: https://forms.gle/TCdGem3WTUYEP31B8

Direct download: false-concensus.mp3
Category:general -- posted at: 3:16pm PDT

In this solo episode, Kyle overviews the field of fraud detection with eCommerce as a use case.  He discusses some of the techniques and system architectures used by companies to fight fraud with a focus on why these things need to be approached from a real-time perspective.

Direct download: fraud-detection-in-real-time.mp3
Category:general -- posted at: 12:12am PDT

In this episode, Kyle and Linhda review the results of our recent survey. Hear all about the demographic details and how we interpret these results.

Direct download: listener-survey-review.mp3
Category:general -- posted at: 10:01am PDT

Moses Namara from the HATLab joins us to discuss his research into the interaction between privacy and human-computer interaction.

Direct download: human-computer-interaction-and-online-privacy.mp3
Category:general -- posted at: 2:43pm PDT




Direct download: authorship-attribution-of-lennon-mccartney-songs.mp3
Category:general -- posted at: 8:00am PDT

Erik Härkönen joins us to discuss the paper GANSpace: Discovering Interpretable GAN Controls. During the interview, Kyle makes reference to this amazing interpretable GAN controls video and it’s accompanying codebase found here. Erik mentions the GANspace collab notebook which is a rapid way to try these ideas out for yourself.

Direct download: gans-can-be-interpretable.mp3
Category:general -- posted at: 7:42pm PDT

Direct download: sentiment-preserving-fake-reviews.mp3
Category:general -- posted at: 3:48pm PDT

Sungsoo Ray Hong joins us to discuss the paper Human Factors in Model Interpretability: Industry Practices, Challenges, and Needs.

Direct download: interpretability-practitioners.mp3
Category:general -- posted at: 9:43am PDT

Deb Raji joins us to discuss her recent publication Saving Face: Investigating the Ethical Concerns of Facial Recognition Auditing.

Direct download: facial-recognition-auditing.mp3
Category:general -- posted at: 11:34am PDT




Direct download: robust-fit-to-nature.mp3
Category:general -- posted at: 8:56am PDT

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 cases, their accuracy makes them incredibly useful.

But does achiving “usefulness” require a black box? Can we be sure an equally valid but simpler solution does not exist?

Cynthia Rudin helps us answer that question. We discuss her recent paper with co-author Joanna Radin titled (spoiler warning)…

Why Are We Using Black Box Models in AI When We Don’t Need To? A Lesson From An Explainable AI Competition




Direct download: black-boxes-are-not-required.mp3
Category:general -- posted at: 12:59pm PDT

Daniel Kang joins us to discuss the paper Testing Robustness Against Unforeseen Adversaries.

Direct download: robustness-to-unforeseen-adversarial-attacks.mp3
Category:general -- posted at: 8:29am PDT



Direct download: estimating-the-size-of-language-acquisition.mp3
Category:general -- posted at: 2:36pm PDT

Jayaraman Thiagarajan joins us to discuss the recent paper Calibrating Healthcare AI: Towards Reliable and Interpretable Deep Predictive Models.

Direct download: interpretable-ai-in-healthcare.mp3
Category:general -- posted at: 8:49am PDT

What does it mean to understand a neural network? That’s the question posted on this arXiv paper. Kyle speaks with Tim Lillicrap about this and several other big questions.



Direct download: understanding-neural-networks.mp3
Category:general -- posted at: 10:07am PDT

Dan Elton joins us to discuss self-explaining AI. What could be better than an interpretable model? How about a model wich explains itself in a conversational way, engaging in a back and forth with the user.

We discuss the paper Self-explaining AI as an alternative to interpretable AI which presents a framework for self-explainging AI.



Direct download: self-explaining-ai.mp3
Category:general -- posted at: 10:23pm PDT

Becca Taylor joins us to discuss her work studying the impact of plastic bag bans as published in Bag Leakage: The Effect of Disposable Carryout Bag Regulations on Unregulated Bags from the Journal of Environmental Economics and Management. How does one measure the impact of these bans? Are they achieving their intended goals? Join us and find out!

Direct download: plastic-bag-bans.mp3
Category:general -- posted at: 8:45am PDT




Direct download: self-driving-cars-and-pedestrians.mp3
Category:general -- posted at: 10:58am PDT

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

Direct download: computer-vision-is-not-perfect.mp3
Category:general -- posted at: 10:53am PDT

Jessica Hullman joins us to share her expertise on data visualization and communication of data in the media. We discuss Jessica’s work on visualizing uncertainty, interviewing visualization designers on why they don't visualize uncertainty, and modeling interactions with visualizations as Bayesian updates.

Homepage: http://users.eecs.northwestern.edu/~jhullman/

Lab: MU Collective

Direct download: uncertainty-representations.mp3
Category:general -- posted at: 8:18am PDT

Announcing Journal Club

I am pleased to announce Data Skeptic is launching a new spin-off show called "Journal Club" with similar themes but a very different format to the Data Skeptic everyone is used to.

In Journal Club, we will have a regular panel and occasional guest panelists to discuss interesting news items and one featured journal article every week in a roundtable discussion. Each week, I'll be joined by Lan Guo and George Kemp for a discussion of interesting data science related news articles and a featured journal or pre-print article.

We hope that this podcast will give listeners an introduction to the works we cover and how people discuss these works. Our topics will often coincide with the original Data Skeptic podcast's current Interpretability theme, but we have few rules right now or what we pick. We enjoy discussing these items with each other and we hope you will do.

In the coming weeks, we will start opening up the guest chair more often to bring new voices to our discussion. After that we'll be looking for ways we can engage with our audience.

Keep reading and thanks for listening!

Kyle

Direct download: AlphaGo_COVID-19_Contact_Tracing_and_New_Data_Set.mp3
Category:general -- posted at: 11:00pm PDT

Direct download: visualizing-uncertainty.mp3
Category:general -- posted at: 8:00am PDT

Pramit Choudhary joins us to talk about the methodologies and tools used to assist with model interpretability.

Direct download: interpretability-tooling.mp3
Category:general -- posted at: 8:00am PDT

Kyle and Linhda discuss how Shapley Values might be a good tool for determining what makes the cut for a home renovation.

Direct download: shapley-values.mp3
Category:general -- posted at: 12:29pm PDT

We welcome back Marco Tulio Ribeiro to discuss research he has done since our original discussion on LIME.

In particular, we ask the question Are Red Roses Red? and discuss how Anchors provide high precision model-agnostic explanations.


Please take our listener survey.

Direct download: anchors-as-explanations.mp3
Category:general -- posted at: 6:46am PDT

Direct download: mathematical-models-of-ecological-systems.mp3
Category:general -- posted at: 4:10pm PDT

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.

Direct download: adversarial-explanations.mp3
Category:general -- posted at: 3:10pm PDT

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/

Direct download: objectnet.mp3
Category:general -- posted at: 8:00am PDT

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!

Direct download: visualization-and-interpretability.mp3
Category:general -- posted at: 8:00am PDT

We welcome Su Wang back to Data Skeptic to discuss the paper Distributional modeling on a diet: One-shot word learning from text only.

Direct download: interpretable-one-shot-learning.mp3
Category:general -- posted at: 9:00pm PDT

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.

Direct download: fooling-computer-vision.mp3
Category:general -- posted at: 10:38am PDT

This episode includes an interview with Aaron Roth author of The Ethical Algorithm.

Direct download: algorithmic-fairness.mp3
Category:general -- posted at: 6:31pm PDT

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.

Direct download: interpretability.mp3
Category:general -- posted at: 12:33am PDT

A year in recap.

Direct download: nlp-in-2019.mp3
Category:general -- posted at: 3:51am PDT

We are joined by Colin Raffel to discuss the paper "Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer".

Direct download: the-limits-of-nlp.mp3
Category:general -- posted at: 5:18pm PDT

Seth Juarez joins us to discuss the toolbox of options available to a data scientist to jumpstart or extend their machine learning efforts.

Direct download: jumpstart-your-ml-project.mp3
Category:general -- posted at: 9:25am PDT

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.

Direct download: serverless-nlp-model-training.mp3
Category:general -- posted at: 6:13pm PDT

Buck Woody joins Kyle to share experiences from the field and the application of the Team Data Science Process - a popular six-phase workflow for doing data science.

 

Direct download: the-team-data-science-process.mp3
Category:general -- posted at: 2:54pm PDT

Thea Sommerschield joins us this week to discuss the development of Pythia - a machine learning model trained to assist in the reconstruction of ancient language text.

Direct download: ancient-text-restoration.mp3
Category:general -- posted at: 10:25pm PDT

Kyle met up with Damian Brady at MS Ignite 2019 to discuss machine learning operations.

Direct download: ml-ops.mp3
Category:general -- posted at: 12:18am PDT

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

 

Direct download: annotator-bias.mp3
Category:general -- posted at: 1:46pm PDT

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.

Direct download: nlp-for-developers.mp3
Category:general -- posted at: 7:00pm PDT

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.

Direct download: indigenous-american-language-research.mp3
Category:general -- posted at: 1:40am PDT

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.

Direct download: talking-to-gpt2.mp3
Category:general -- posted at: 12:45pm PDT

Rajiv Shah attempted to reproduce an earthquake-predicting deep learning model.  His results exposed some issues with the model.  Kyle and Rajiv discuss the original paper and Rajiv's analysis.

Direct download: reproducing-deep-learning-models.mp3
Category:general -- posted at: 6:15pm PDT

Allyson Ettinger joins us to discuss her work in computational linguistics, specifically in exploring some of the ways in which the popular natural language processing approach BERT has limitations.

Direct download: what-bert-is-not.mp3
Category:general -- posted at: 2:02pm PDT

Omer Levy joins us to discuss "SpanBERT: Improving Pre-training by Representing and Predicting Spans".

https://arxiv.org/abs/1907.10529

Direct download: spanbert.mp3
Category:general -- posted at: 1:27am PDT

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.

Direct download: bert-is-shallow.mp3
Category:general -- posted at: 2:13pm PDT

Kyle pontificates on how impressed he is with BERT.

Direct download: bert-is-magic.mp3
Category:general -- posted at: 10:11pm PDT

Kyle sits down with Jen Stirrup to inquire about her experiences helping companies deploy data science solutions in a variety of different settings.

Direct download: applied-data-science-in-industry.mp3
Category:general -- posted at: 10:31pm PDT

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

Direct download: building-the-howto100m-video-corpus.mp3
Category:general -- posted at: 1:12pm PDT

Kyle provides a non-technical overview of why Bidirectional Encoder Representations from Transformers (BERT) is a powerful tool for natural language processing projects.

Direct download: bert.mp3
Category:general -- posted at: 11:42pm PDT

Kyle interviews Prasanth Pulavarthi about the Onnx format for deep neural networks.

Direct download: onyx.mp3
Category:general -- posted at: 12:52am PDT

Kyle and Linhda discuss some high level theory of mind and overview the concept machine learning concept of catastrophic forgetting.

Direct download: catastrophic-forgetting.mp3
Category:general -- posted at: 1:40am PDT

Sebastian Ruder is a research scientist at DeepMind.  In this episode, he joins us to discuss the state of the art in transfer learning and his contributions to it.

Direct download: transfer_learning.mp3
Category:general -- posted at: 9:02pm PDT

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.

Direct download: facebook-language.mp3
Category:general -- posted at: 9:21am PDT

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.

Direct download: under-resourced-languages.mp3
Category:general -- posted at: 3:17pm PDT

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).

Direct download: named-entity-recognition.mp3
Category:general -- posted at: 11:16am PDT

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.

Direct download: the-death-of-a-language.mp3
Category:general -- posted at: 2:47pm PDT

Kyle and Linh Da discuss the concepts behind the neural Turing machine.

Direct download: neuro-turing-machines.mp3
Category:general -- posted at: 9:05am PDT

Kyle chats with Rohan Kumar about hyperscale, data at the edge, and a variety of other trends in data engineering in the cloud.

Direct download: data-infrastructure-in-the-cloud.mp3
Category:general -- posted at: 12:28pm PDT

In this episode, Kyle interviews Laura Edell at MS Build 2019.  The conversation covers a number of topics, notably her NCAA Final 4 prediction model.

 

Direct download: ncaa-predictions-on-spark.mp3
Category:general -- posted at: 9:52am PDT

Kyle and Linhda discuss attention and the transformer - an encoder/decoder architecture that extends the basic ideas of vector embeddings like word2vec into a more contextual use case.

Direct download: transformer.mp3
Category:general -- posted at: 8:31am PDT

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.

 

Direct download: mapping-dialects-with-twitter-data.mp3
Category:general -- posted at: 8:00am PDT

This is an interview with Ellen Loeshelle, Director of Product Management at Clarabridge.  We primarily discuss sentiment analysis.

Direct download: sentiment-analysis.mp3
Category:general -- posted at: 6:46pm PDT

A gentle introduction to the very high-level idea of "attention" in machine learning, as it will play a major role in some upcoming episodes over the next few weeks.

Direct download: attention-part-1.mp3
Category:general -- posted at: 7:46pm PDT

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.

 

Direct download: cross-lingual.mp3
Category:general -- posted at: 6:42am PDT

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.

Direct download: elmo.mp3
Category:general -- posted at: 8:00am PDT

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?

Direct download: bleu.mp3
Category:general -- posted at: 9:16pm PDT

While at NeurIPS 2018, Kyle chatted with Liang Huang about his work with Baidu research on simultaneous translation, which was demoed at the conference.

Direct download: simultaneous-translation.mp3
Category:general -- posted at: 8:00am PDT

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!

Direct download: human-vs-machine-transcription-errors.mp3
Category:general -- posted at: 8:00am PDT

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

Direct download: seq2seq.mp3
Category:general -- posted at: 8:00am PDT

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

Direct download: text-mining-in-r.mp3
Category:general -- posted at: 8:00am PDT

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

Direct download: recurrent-relational-networks.mp3
Category:general -- posted at: 7:47am PDT

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.

Direct download: text-world-and-word-embedding-lower-bounds.mp3
Category:general -- posted at: 8:00am PDT

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.

Direct download: word2vec.mp3
Category:general -- posted at: 8:00am PDT

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.

Direct download: authorship-attribution.mp3
Category:general -- posted at: 8:44am PDT

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.

Direct download: extremely-large-corpora.mp3
Category:general -- posted at: 8:00am PDT

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.

Direct download: semantic-search-at-github.mp3
Category:general -- posted at: 8:00am PDT

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/

 

 

Direct download: natural-language-processing.mp3
Category:general -- posted at: 8:15am PDT

Kyle shares a few thoughts on mistakes observed by job applicants and also shares a few procedural insights listeners at early stages in their careers might find value in.

Direct download: data-science-hiring-processes.mp3
Category:general -- posted at: 8:00am PDT

Epicac by Kurt Vonnegut.

Direct download: epicac.mp3
Category:general -- posted at: 10:07pm PDT

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.

Direct download: drug-discovery.mp3
Category:general -- posted at: 8:00am PDT

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.

Direct download: sign-language-recognition.mp3
Category:general -- posted at: 8:00am PDT

 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.

Direct download: data-ethics.mp3
Category:general -- posted at: 8:00am PDT

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.

Direct download: escaping-the-rabbit-hole.mp3
Category:general -- posted at: 9:06am PDT

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.

 

Direct download: theorem-provers.mp3
Category:general -- posted at: 8:09am PDT

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.

Direct download: automated-fact-checking.mp3
Category:general -- posted at: 8:45am PDT

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.

Direct download: single-source-of-truth.mp3
Category:general -- posted at: 7:00am PDT

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.

Direct download: detecting-fast-radio-bursts-with-deep-learning.mp3
Category:general -- posted at: 9:38am PDT

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.

Direct download: bayesian-redux.mp3
Category:general -- posted at: 8:00am PDT

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.

Direct download: modeling-fake-news.mp3
Category:general -- posted at: 8:00am PDT

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 [-1, 1] 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.

Direct download: louvain-community-detection.mp3
Category:general -- posted at: 8:22am PDT

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.

Direct download: cultural-cognition.mp3
Category:general -- posted at: 8:24am PDT

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. 

Direct download: false-discovery-rates.mp3
Category:general -- posted at: 8:09am PDT

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:

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.

Direct download: deepfakes.mp3
Category:general -- posted at: 8:00am PDT