Microsoft Research posted this video of Sumit Gulwani, who founded the PROSE research and engineering team at Microsoft. This team develops programming-by-example (PBE) APIs and ships them through multiple Microsoft products.
PBE is a new frontier in AI wherein the computer programs itself—the user provides input-output examples and the computer synthesizes an intended script. This is significant because 99% of computer users do not know programming. Even for programmers, this can provide a 10-100x productivity increase for many task domains.
A killer application of PBE is in the space of data cleaning/preparation since data scientists often spend up to 80% time wrangling data into a form suitable for learning models or drawing insights. In this video, Sumit illustrates how a data cleaning task, that Python programmers took an average of 30 minutes to finish, can be performed in 30 seconds by non-programmers using the PBE paradigm. In particular, PBE can help ingest a file into tabular format, split a column to extract constituent sub-fields, derive new columns, and suggest form entries.
TensorFlow is a powerful data flow-oriented machine learning framework developed by Google’s Brain Team. It was designed to be easy to use and widely applicable on both numeric, neural network-oriented problems as well as other domains.
In this video, Siraj Raval explains the “DeepFakes” algorithm and phenomenon, as well as start us thinking about to deal with fraudulent AI generated content.
In case you’re curious about how Convolutional Neural Networks work in regards to image recognition, but were scared off by the math, then here’s the video for you.
Microsoft Research has just posted this video. Don’t worry about the poor sound quality in the beginning, it gets better when the main presenter starts speaking.
Siraj Raval has a unique and innovative idea on how to finance his “School of AI.”
Two Minute Papers examines the paper “Autonomous Reconstruction of Unknown Indoor Scenes Guided by Time-varying Tensor Fields” and it’s impressive.
From time to time, the term “non-Euclidian geometry” comes up in the context of creating models for AI systems.
Here’s a great explainer of what non-Euclidian geometry.
Last week, I blogged a video by CGP Grey about how machines really learn. While the explanation was accurate, succinct, and beautiful it didn’t cover all the ways that a machine could learn.
This video explains a few more ways that machines learn.