(Automated) Curriculum Learning

- lmn

I’ve lately spent some time reading about Curriculum Learning and experimenting with the algorithms described in two of the papers in this domain

Bengio, Y., Louradour, J., Collobert, R., & Weston, J. (2009, June). Curriculum learning. In Proceedings of the 26th annual international conference on machine learning (pp. 41-48). ACM.

Graves, A., Bellemare, M. G., Menick, J., Munos, R., & Kavukcuoglu, K. (2017). Automated Curriculum Learning for Neural Networks. arXiv preprint arXiv:1704.03003.

The first of the above can be considered important given how with empirical results supporting Curriculum Learning, it revived the interest among researchers in this technique. The second is one of the recently proposed approaches for Curriculum Learning that I thought would be interesting to understand in greater depth.

I’ve summarised my thoughts on these in a short presentation. I hope to share my code and results not too long from now as well.

Completed the Course “Introduction to Big Data” offered by UCSD on Coursera

- lmn

I successfully completed this course with a 98.9% mark. It was easy and covered mostly definitions, some history of big data, big data jargon and very basic principles. There was an emphasis on what constitutes big data (in terms of size, variety, complexity, etc.), what kinds of analyses one can carries out on big data, what sources they can be from, and what tools one could use to analyse them. When it came to the latter, the course offered a brief introduction to the Hadoop  ecosystem that I found particularly interesting as I hadn’t ever worked with any of the software that is a part of this ecosystem. And there was also a simple assignment that gave one a taste of what working with Hadoop could be like. Here’s a link to the certificate I received from Coursera on completing this course.

Looking forward to the remaining courses in the Big Data specialisation!

Completed Andrew Ng’s “Improving Deep Neural Networks” course on Coursera

- lmn

I successfully completed this course with a 100.0% mark. Once again, this course was easy given my experience so far in machine learning and deep learning. However, as with the previous course I completed in the same specialisation there were a few things that were worth attending this course for. I particularly found the sections on Optimisation (exponential moving averages, Momentum, RMSProp and Adam optimisers, etc.), Batch Normalisation, and to some extend Dropout useful. Here’s a link to the certificate from Coursera for this course.

I’m looking forward to the course on Convolutional Neural Networks!

Invited Talks at the International Institute of Information Technology – Bangalore and Robert Bosch

- lmn

I’m currently on a break from work at Jukedeck until the 22nd of September, and visiting friends and old colleagues in Bangalore for a few days. On coming to know of my visit to Bangalore, my past mentors invited me to give talks at their respective organisations – the International Institute of Information Technology – Bangalore, and Robert Bosch. Today I presented the work I did on sequence modelling in music, RBMs and Recurrent RBMs during my PhD to the staff and students at the International Institute of Information Technology – Bangalore (IIIT-B). And next Monday (the 18th of September, 2017) it will be more or less the same talk at Robert Bosch.

Here is a copy of the slides for those presentations.

Participating in CSMC 2017 Panel Discussion

- lmn

At 11:30 on the 13th of September, 2017  I will be participating in a panel discussion on the subject of “Applying Musical Patterns in Generation” together with Elaine Chew, Roger Dean, Steven Jan, David Meredith and Tillman Weyde. It is being organised by Iris Yuping Ren as a part of the 2nd Conference on Computer Simulation of Musical Creativity between the 11th-13th of September, 2017 at Milton-Keynes, UK.

Really excited and looking forward to it!

Completed Andrew Ng’s “Structuring Machine Learning Projects” course on Coursera

- lmn

I successfully completed this course with a 96.7% mark. It was fairly easy given my experience so far in machine learning and deep learning, but there were a few new ideas that I learned here and also others that I investigated in greater depth out of my own curiosity while doing it. I felt like the Transfer Learning, Multitask Learning and End-to-End ML lectures are not really useful immediately after the course unless one takes these up after the course in greater depth as the lectures on these topics were quite superficial and brief. The practical advice, however, and the hand-on exercises that focused on real-world scenarios were useful and I wish there was more of the latter (perhaps optional) in the course.

Here’s a link to the certificate I received from Coursera for this course.

My Experience in Applying for a Work Visa at the UK Home Office

- lmn

This is more a rant out of frustration than anything else, and I hope this will help others get a sense of what a nightmare it can be to deal with the UK Home Office when something goes wrong.

Over 2 months ago on the 9th of May, 2017 I submitted an application for a Tier 2 Work visa with my employer’s backing to the UK Home Office. With the fee that we paid for this application (GBP 1,354.00), a processing time of 8 weeks or less was guaranteed to us on the Home Office website. However, it has been nearly 10 weeks now and we have not received a decision or a status update on my visa application. I called them up around 2 weeks ago when the 8-week period had passed (5th of July, 2017) and after a near 30 minute wait, the lady who answered my call casually told me that there were delays, that I would hear from them in “a couple of weeks”, and not to panic. I explained to her how the fact that the Home Office is in possession of my passport and current residence permit is causing me a lot of inconvenience and she recommended that I request my documents back from them through their website. I tried this, but I was not allowed to do so without withdrawing my application altogether as my employer is not a “Premium Sponsor” – they’re only a young start-up so I was kind of expecting this to be the case.

A few days after this, I came to know through the Citizen’s Advice website that I might be able to get more information about the progress of my application or possibly have it expedited by contacting my local MP (Ms. Marsha de Cordova who is the Labour MP for Battersea) and I did that as well. I’m now waiting for her office to respond. In the meantime, I checked the status of my application on the Home Office website on Saturday (15th July, 2017) and was surprised to see that a decision on my application had been made on the 4th of June, 2017 and so I should have received my documents back by the 14th of June. Neither of those have happened and we’re still in the dark as to the status of my application and the whereabouts of my passport and current residence permit.

I, once again, called the Home office earlier today (17th of July, 2017) to inquire about my application in light of the aforementioned new information that a decision had been made about it. After nearly an hour long wait, I got through to a representative. All the service he said that he could offer me at that point was to forward the details of my application to one of his colleagues who I would hear back from in 3-5 working days. I was also struck by his lack of any empathy whatsoever when I expressed my concern and anxiety on being kept in the dark about the state of my application and important documents way beyond the service standard that was communicated to me. This is besides the point anyway – I can’t expect some random representative of the Home Office to act as my crying shoulder. I hung up feeling a bit worthless, but whatever.

I’m very unhappy about this entire experience. And it seems to defeat the purpose of making a formal complaint about the Home Office processes to a department in the Home Office itself but I’ll do it anyway. My passport being in possession of the Home Office beyond the deadline for processing my application is inconvenient, to say the least. I have had to refrain from any travel outside the UK, and hold off making financial transactions between the UK and India through my bank as my passport is required for these purposes. I could prepare myself to put up with it for the two months that I was told it would take to process my application, however, now I have absolutely no idea when my documents would be returned to me, where they are, and when a decision on my visa application will be made. I found the response from the Home Office representatives very unsatisfactory, and I feel that I am being taken for granted by being kept in the dark with no sense of urgency in returning my documents.

I’ll wait for the 3-5 days as I’ve been told. If anyone who reads this has any other advice for me that might help me, please do post your advice in the comments below. I would appreciate it.

Edit (25th of July, 2017): Got my passport back with the approval and the new residence permit as well. As it turns out, these were mailed to an address that I moved out of in mid-May after making the application when I was still living there. To their credit, the Home Office did indeed do the job well within time (and I really appreciate the caseworker’s effort when in comes to that). I do still maintain that their helpdesk is by far the worst that I have ever come across with unsympathetic representatives, incredibly long waiting time, and their inability to give me simple answers regarding my application. Also, having had access to both my email address and telephone number, it would’ve helped if they had communicated the fact that my application had been processed through at least one of these channels instead of relying solely on Royal Mail.

Passed Rock School Grade 6 (Electric Guitar) Exam

- lmn

A couple of weeks ago on Jun 9, 2017 I gave my first music certification exam – RSL Awards Rock School Grade 6 (Electric Guitar). I’m very pleased to share that I passed it with a distinction! Before the exam, I recorded my performances of the three pieces I chose to play as a part of it. The entire piece is given except the solos which have been composed by me. I uploaded them on YouTube, initially to share with my tutor, and now with everyone else! Here they are…

D’oro

- lmn

I like using the Pomodoro Technique to remind me to take short breaks while I’m immersed in work. And I thought it would be nice to create a little command-line Pomodoro Timer for myself that will pop up desktop notifications telling me that it’s time to take a break. I call this very simple and minimal Pomodoro timer app D’oro and it can be invoked using a command called doro once it has been installed.

You can clone the repository from its Gitlab page . I intend to write an installation script and also create a Debian package for it in the future, but it works and I love using it everyday at work!

Breaking Down the Differentiable Neural Computer

- lmn

Ever since first working with Recurrent Neural Networks (RNNs) for predicting musical sequences during my PhD, I have been fascinated by these models and try to keep up with exciting developments in connectionist machine learning research surrounding these models. One of these for me has been the emergence of RNNs that are augmented by a dedicated memory unit. The idea was notably illustrated as the Neural Turing Machine (NTM) in an ArXiV submission by Alex Graves and colleagues from Google DeepMind. This early work while having gathered a fair deal of acclaim in the community, has since been followed up in a publication in the prestigious journal Nature that introduces a more evolved variant of the NTM known as the Differentiable Neural Computer (DNC). During the past couple of weeks, I managed to spend some time learning about the NTM and the DNC and prepared a little slide-show (with Google Slides) containing my observations to share with others.

So here is the link to the slides, and I hope some of you who read it benefit from it! Please let me know if you find anything that needs to be corrected in it. I would appreciate that!