[Links of the Day] 28/06/2018 : Operational CRDT & causal trees, the story ISPC and Larabee compiler, Limitation of gradient descent

• Causal Trees& Operational CRDTs : Educational project showing how to use CRDT for real-time document sharing and editing.
• The story of ispc : Intel Larrabee compiler history. It seems that Intel missed the mark there, and was significantly too early for the deep learning onslaught. It seems that to some extent that the ISPC model would have significantly bridged the gap between GPU and CPU for machine learning computation. 
• The limitations of gradient descent as a principle of brain function : looks like emulating more complex brain function will not work by using gradient descent methods. While this strategy was quite successful for deep learning it seems that there is some inherent limitation to a more generic brain functions emulation as the authors describe. 

[Links of the Day] 26/06/2018 : How economist got Brexit wrong, Driving data set, CRDT @ redis

• How the economics profession got it wrong on Brexit : Economist got the economy wrong... News at 11 .. Anyway, it's a very good analyse of the pitfalls that the various group fell into. And a good read to get a better understanding of the UK economy and how to reacts to large socio-economic events. 
• BDD100K : want data for your driverless car ?? Berkeley got you covered.  [data][paper]
• CRDT @ redis : I love CRDT and this talk about their use in Redis.

[Links of the Day] 16/01/2018 : planetary scale DB - AntidoteDB, Benchmarks for Machine Learning and the hardware running the algorithms

• AntidoteDB : large scale ( planet-scale ) distributed DB system. Competing with the like of cockroachDB or spanner. The core differentiator the architecture heavily rely on CRDT for its core functionality. It is a spin-off from the SyncFree EU research project. Sadly like a lot of EU or research-driven startup spin-off the documentation and website are slightly lacking polish. The architecture reference link is broken and a lot of stuff seems to be work in progress. Common guys! If you want to build a community and a product you really need to pick up the pace. This project has great potential, don't let it go to waste. 
• Machine Learning Benchmarks - Hardware Provider : a very good survey of machine learning benchmark of the current cloud provider. What is even more useful from that benchmark is that you get a cost overview of running ML application. Which is often a big unknown at the moment. 
• DeepMind Control Suite : benchmark suite for machine learning algorithms using a set of continuous control tasks with a standardised structure and interpretable rewards

[Links of the Day] 31/10/2017 : Machine learning at the edge, Deep Learning on Hardware , operations based CRDT

• EdgeML : Microsoft research demonstrate how to push machine learning at the edge and run KB models. We could quickly see machine learning enabled IoT device popping around us.  [slides] [github]
• Efficient Methods and Hardware for Deep Learning : Quest for speed never stop, and often that means getting read of those pesky indirection layers that make your software architecture so flexible :) 
• Pure Operation-Based Replicated Data Types : CRDT for operations rather than just value. But the core concepts are a little bit tricky and there is some potential pitfall in the approach. Such as the performance limitation and the reliance on causal stability ( which is really hard to obtain in pure decentralised systems)

[Links of the day] 23/03/2016: containers patterns, delta CRDTs, probabilistic DB

• Container Patterns : WiP but promising documentation of containers patterns. Check v1.0 branch 
• Efficient State-based CRDTs by Delta-Mutation :  instead of maintaining a full information in a CRDT the authors propose to use delta based messages in order to reduce storage and network space overhead.
• BlinkDB : allows users to trade-ošff query accuracy for response time, enabling interactive queries over massive data by running queries on data samples and presenting results annotated with meaningful error bars. Really cool, we start to see the emergence of probabilistic programming everywhere. We just have to get used to that like real life, computer programs can be more efficient when not everything is certain.

Links of the day 11 - 03 - 2015

Today's links 11/03/2015: category theory, linux profiling, CRDT, Log structured merge trees

• Category Theory : notes and list of article, books on category theory.
• Linux Profiling at Netflix : perf and flame-graph, should be part of every body standard toolbox. 
• Consistency without Consensus: CRDTs in Production at SoundCloud
• Log Structured Merge Trees : Instead of having one big index structure (which will either scatter-gun the file system or add significant write amplification) batches of writes are saved, sequentially, to a set of smaller index files.