Learning to prove theorems via interacting with proof assistants Yang & Deng, ICML'19 Something a little different to end the week: deep learning meets theorem proving! It’s been a while since we gave formal methods some love on The Morning Paper, and this paper piqued my interest. You’ve probably heard of Coq, a proof management … Continue reading Learning to prove theorems via interacting with proof assistants
Author: adriancolyer
Venture Partner with Accel in London, author of 'The Morning Paper'
Statistical foundations of virtual democracy
Statiscal foundations of virtual democracy Kahng et al., ICML'19 This is another paper on the theme of combining information and making decisions in the face of noise and uncertainty - but the setting is quite different to those we’ve been looking at recently. Consider a food bank that receives donations of food and distributes it … Continue reading Statistical foundations of virtual democracy
Robust learning from untrusted sources
Robust learning from untrusted sources Konstantinov & Lampert, ICML'19 Welcome back to a new term of The Morning Paper! Just before the break we were looking at selected papers from ICML’19, including “Data Shapley.” I’m going to pick things up pretty much where we left off with a few more ICML papers... Data Shapley provides … Continue reading Robust learning from untrusted sources
End of term
I can't believe we've arrived at the end-of-term again already! I'll be taking a four-week break from writing The Morning Paper, normal service resumes on Monday 19th August. A big milestone will slip quietly by during this recess - it was five years ago on the 30th July 2014 that I read and shared the … Continue reading End of term
Meta-learning neural Bloom filters
Meta-learning neural bloom filters Rae et al., ICML'19 Bloom filters are wonderful things, enabling us to quickly ask whether a given set could possibly contain a certain value. They produce this answer while using minimal space and offering O(1) inserts and lookups. It’s no wonder Bloom filters and their derivatives (the family of approximate set … Continue reading Meta-learning neural Bloom filters
Challenging common assumptions in the unsupervised learning of disentangled representations
Challenging common assumptions in the unsupervised learning of disentangled representations Locatello et al., ICML'19 Today’s paper choice won a best paper award at ICML’19. The ‘common assumptions’ that the paper challenges seem to be: "unsupervised learning of disentangled representations is possible, and useful!" The key idea behind the unsupervised learning of disentangled representations is that … Continue reading Challenging common assumptions in the unsupervised learning of disentangled representations
Data Shapley: equitable valuation of data for machine learning
Data Shapley: equitable valuation of data for machine learning Ghorbani & Zou et al., ICML'19 It’s incredibly difficult from afar to make sense of the almost 800 papers published at ICML this year! In practical terms I was reduced to looking at papers highlighted by others (e.g. via best paper awards), and scanning the list … Continue reading Data Shapley: equitable valuation of data for machine learning
View-centric performance optimization for database-backed web applications
View-centric performance optimization for database-backed web applications Yang et al., ICSE 2019 The problem set-up in this paper discusses the importance of keeping web page load times low as a fundamental contributor to user satisfaction (See e.g. ‘Why performance matters’). Between client-side tools such as Google’s Lighthouse, back-end tools that can analyse ORM usage and … Continue reading View-centric performance optimization for database-backed web applications
Three key checklists and remedies for trustworthy analysis of online controlled experiments at scale
Three key checklists and remedies for trustworthy analysis of online controlled experiments at scale Fabijan et al., ICSE 2019 Last time out we looked at machine learning at Microsoft, where we learned among other things that using an online controlled experiment (OCE) approach to rolling out changes to ML-centric software is important. Prior to that … Continue reading Three key checklists and remedies for trustworthy analysis of online controlled experiments at scale
Software engineering for machine learning: a case study
Software engineering for machine learning: a case study Amershi et al., ICSE'19 Previously on The Morning Paper we’ve looked at the spread of machine learning through Facebook and Google and some of the lessons learned together with processes and tools to address the challenges arising. Today it’s the turn of Microsoft. More specifically, we’ll be … Continue reading Software engineering for machine learning: a case study
the morning paper 