Unsupervised learning of artistic styles with archetypal style analysis Wynen et al., NeurIPS'18 I’ve always enjoyed following work on artistic style transfer. The visual nature makes it easy to gain an appreciation for what is going on and the results are very impressive. It also something that’s been unfolding within the timespan of The Morning … Continue reading Unsupervised learning of artistic styles with archetypal style analysis
Author: adriancolyer
Venture Partner with Accel in London, author of 'The Morning Paper'
Neural Ordinary Differential Equations
Neural ordinary differential equations Chen et al., NeurIPS'18 ‘Neural Ordinary Differential Equations’ won a best paper award at NeurIPS last month. It’s not an easy piece (at least not for me!), but in the spirit of ‘deliberate practice’ that doesn’t mean there isn’t something to be gained from trying to understand as much as possible. … Continue reading Neural Ordinary Differential Equations
The tradeoffs of large scale learning
The tradeoffs of large scale learning Bottou & Bousquet, NIPS'07 Welcome to another year of The Morning Paper. As usual we’ll be looking at a broad cross-section of computer science research (I have over 40 conferences/workshops on my list to keep an eye on as a start!). I’ve no idea yet what papers we’ll stumble … Continue reading The tradeoffs of large scale learning
Towards a theory of software development expertise
Towards a theory of software development expertise Baltes et al., ESEC/FSE'18 This is the last paper we’ll be looking at this year, so I’ve chosen something a little more reflective to leave you with (The Morning Paper will return on Monday 7th January, 2019). The question Baltes and Diehl tackle is this: "How do you … Continue reading Towards a theory of software development expertise
Identifying impactful service system problems via log analysis
Identifying impactful service system problems via log analysis He et al., ESEC/FSE'18 If something is going wrong in your system, chances are you’ve got two main sources to help you detect and resolve the issue: logs and metrics. You’re unlikely to be able to get to the bottom of a problem using metrics alone (though … Continue reading Identifying impactful service system problems via log analysis
Applied machine learning at Facebook: a datacenter infrastructure perspective
Applied machine learning at Facebook: a datacenter infrastructure perspective Hazelwood et al., _HPCA’18 _ This is a wonderful glimpse into what it’s like when machine learning comes to pervade nearly every part of a business, with implications top-to-bottom through the whole stack. It’s amazing to step back and think just how fundamentally software systems have … Continue reading Applied machine learning at Facebook: a datacenter infrastructure perspective
Darwinian data structure selection
Darwinian data structure selection Basios et al., FSE'18 GraphIt may have caught your attention for the success of its approach, but I suspect for many readers it’s not something you’ll be immediately applying. Darwinian Data Structures (DDSs) on the other hand looks to be of immediate interest to many Java and C++ projects (and generalises … Continue reading Darwinian data structure selection
GraphIt: A high-performance graph DSL
GraphIt: a high-performance graph DSL Zhang et al., OOPSLA'18 See also: http://graphit-lang.org/. The problem with finding the optimal algorithm and data structures for a given problem is that so often it depends. This is especially true when it comes to graph algorithms. It is difficult to implement high-performance graph algorithms. The performance bottlenecks of these … Continue reading GraphIt: A high-performance graph DSL
MadMax: surviving out-of-gas conditions in Ethereum smart contracts
MadMax: surviving out-of-gas conditions in ethereum smart contracts Grech et al., OOPSLA'18 We’re transitioning to look at a selection of papers from the recent OOPSLA conference this week. MadMax won a distinguished paper award, and makes a nice bridge from the CCS blockchain papers we were looking at last week. Analysis and verification of smart … Continue reading MadMax: surviving out-of-gas conditions in Ethereum smart contracts
RapidChain: scaling blockchain via full sharding
RapidChain: scaling blockchain via full sharding Zamani et al., CCS'18 RapidChain is a sharding-based public blockchain protocol along the lines of OmniLedger that we looked at earlier in the year. RapidChain is resilient to Byzantine faults from up to 1/3 of its participants, requires no trusted setup, and can achieve more than 7,300 tx/sec with … Continue reading RapidChain: scaling blockchain via full sharding
the morning paper 