About a year and a half ago I saw Jetpac’s video of their Spotter app and I remember thinking at the time that it would be so cool to get this flying on a drone. I didn’t have the bandwidth to work on it at the time but ended up poking at it with Markus Aschinger at the ASI and with two A level students (Jawad / Isaac) from the Nuffield Foundation. While they did good work and it got me a step closer, it still hadn’t quite come together. Hence I sat down the past week to do a full rewrite, integrate it with a quad I had lying around and do a little demo. The result can be seen in the video below.
One of the projects that has taken up a lot of my time the past few months is that of a UAV (drone) based Ground Penetrating Radar (GPR) system. There are a number of applications for this but the one we have been focussing on initially is landmine and UXO clearance. The elements that make up such a system are quite broad. Ranging from sensor design, UAV integration, positioning, terrain following to data analysis. As with many drone projects most of the attention tends to go to the hardware and the flying. While that is certainly important and I have been working on those elements too, the whole system is only as good as the quality and interpretability of the data you get back. That is key. With this post I’ll aim to give a brief summary of the work I have been leading on this front.
There has been a lot going on recently and I thought I would give an update of the main projects I have been working on. I will aim to post more detailed updates and results as things progress and confidentiality agreements allow.
Its late and I’m on my way back from the final demo evening of the Advanced Skills Initiative (ASI). The ASI programme takes disgruntled academics and puts them through an intensive 8 week data science course so they can move into the bustling data science job market. I was a mentor on the machine learning front and it was great to see the fellows evolve and battle through their projects.
One thing struck me though. Something that has been bugging me for a while. Projects pitched at events and meetups like this invariably cover product recommendation, preference learning, financial NLP, churn prediction, adclick prediction, social media trending analysis, etc.
As somebody who likes tangible things, I can’t help but wonder. Where are the projects & startups from aerospace, automotive, marine and other engineering disciplines? Why aren’t we seeing projects from Airbus on smart IVHM, from Jaguar on crumple zones, from Reaction Engines on engine performance, from ASV on marine autopilots, from Princess on vibration control, from McLaren on race strategy, from Dyson on path planning, etc. I personally find these things hugely interesting and fascinating topics. The possibilities for data science / machine learning are endless, prognostics being the obvious example.
Update: These courses have now been taught successfully. If you are interested let me know.
Note: These were originally delivered through Persontyle, however I am no longer affiliated with them in any way after a very unprofessional experience.
I am happy to announce I will be teaching two Machine Learning courses over the summer: A one day high level overview of machine learning is all about and a longer 5 day introductory course that goes into more depth and includes lots of hands on labs with Python scikit-learn. See below for details.
Update: Unfortunately I purposely withdrew from the DL meetup after a very unprofessional experience with the co-organisers Persontyle. I Instead I joined forces with the London Machine Learning Meetup.
It has been almost half a year since I announced I would take over the London Big-O Algorithms meetup and bring it back to life. 5 Months later I am very happy to say all has gone extremely well, interest and attendance far exceeding my expectations. We have had a great set of meetups so far and all speaker slots are booked until June with talks from Google, The founder of ZeroMQ, Microsoft Bing, and many others. Really nice to see there is strong interest in good, solid, technical content.
However, enough about Big-O. This post is to announce a new meetup group that I have been convinced into setting up. The Deep Learning London meetup. The aim of this group is to bring together people interested in the family of machine learning methods that are concerned with learning distributed, hierarchical (“deep”) representations. Neural Networks being the most popular implementation. Its an area I have been looking at for a while and will be getting into quite deeply over the next couple of months (no pun intended). The format will be based around guest speakers sharing new research ideas and applications covering a wide range of fields from computer vision and natural language processing to autonomous systems and prognostics.
Note we are not assuming deep learning is the be-all end-all silver bullet of machine learning and welcome critical thoughts and benchmarks.
Sound interesting? Get in touch!
…to find an algorithm that predicts whether (and for what reason) a question will be closed. The idea is simple: we’ve prepared a dataset with all the questions on Stack Overflow, including everything we knew about them right before they were posted, and whether they finally ended up closed or not. You grab the data, build your brilliant classifier, run it against some leaderboard data and submit your results. Rinse and repeat until the contest ends, when we’ll grab the most promising classifiers and run them against fresh data to choose winners.