pinboard December 11, 2016

• [1602.04938] "Why Should I Trust You?": Explaining the Predictions of Any Classifier
  Despite widespread adoption, machine learning models remain mostly black boxes. Understanding the reasons behind predictions is, however, quite important in assessing trust, which is fundamental if one plans to take action based on a prediction, or when choosing whether to deploy a new model. Such understanding also provides insights into the model, which can be used to transform an untrustworthy model or prediction into a trustworthy one. In this work, we propose LIME, a novel explanation technique that explains the predictions of any classifier in an interpretable and faithful manner, by learning an interpretable model locally around the prediction. We also propose a method to explain models by presenting representative individual predictions and their explanations in a non-redundant way, framing the task as a submodular optimization problem. We demonstrate the flexibility of these methods by explaining different models for text (e.g. random forests) and image classification (e.g. neural networks). We show the utility of explanations via novel experiments, both simulated and with human subjects, on various scenarios that require trust: deciding if one should trust a prediction, choosing between models, improving an untrustworthy classifier, and identifying why a classifier should not be trusted.
• Just say NO to Paxos overhead: replacing consensus with network ordering | the morning paper
• Russian Hackers Acted to Aid Trump in Election, U.S. Says – The New York Times
• Donald Trump Is Gaslighting America | Teen Vogue
• Gareth Edwards on Rogue One, Diversity, Reshoots — Vulture
• When Is Rogue One Set, and Other Questions — Vulture
• Twitter
  My only plan is- don’t die until I’m 130, so I can make some headway on them
• Best Books 2016
• A Cassandra in Trumpland: Sarah Kendzior’s Pithy Commentary on Privilege
• Dissecting the #PizzaGate Conspiracy Theories – The New York Times
• As Aleppo burns in this age of lies, Picasso’s Guernica still screams the truth about war | Art and design | The Guardian
• Dorothea Lange’s Censored Photographs of FDR’s Japanese Concentration Camps — Anchor Editions
• Rex Tillerson, a Candidate for Secretary of State, Has Ties to Vladimir Putin – WSJ
  RT @NateSilver538: Just leavin’ this here:
• Twitter
  RT @NateSilver538: Just leavin’ this here:
• Lions are the Brainiest of the Big Cats – Scientific American
• Gregory Hines & Mikhail Baryshnikov: "Duo Dance" (White Nights 1985) [HD] – YouTube
• Politwoops – Deleted tweets from politicians
• Twitter
  RT @derekwillis: Reminder: you can see deleted tweets from politicians at Politwoops!
• Cindy Gallop on Twitter: "@moorehn Tags here: @ev @jack @marthalanefox @iamdebralee @omidkordestani @peterfenton @marjscar @hughjohnston @rosenblattdavid @btaylor"
• Google’s self-driving car team is hiring executives as it prepares to spin out from Alphabet’s X – Recode
• Trump Isn’t Rewarding a Lot of Loyalists with Cabinet Jobs. Here’s Why. – The New York Times
• Log In – New York Times
  RT @AJentleson: I’ll say it: NYT interviewed Reid for this story. He said things contrary to the story. NYT discarded the interview.
• Using Cloud Dataflow to run parallel predictions with your TensorFlow model – めもめも
• Promoting Open Science in the University (OSBD 2016) // Speaker Deck
• GitHub – jakevdp/PythonDataScienceHandbook: Jupyter Notebooks for the Python Data Science Handbook
• Publishing with GitHub Pages, now as easy as 1, 2, 3 · GitHub
• [1612.02485] Comparative Evaluation of Big-Data Systems on Scientific Image Analytics Workloads
• Presentations by Jake VanderPlas // Speaker Deck
• Twitter

Digest powered by RSS Digest