Travel

Attending the 20th International Society for Music Information Retrieval Conference (ISMIR 2019)

- lmn

It’s just been confirmed that four of us from Moodagent – Reinier de Valk, Pierre Lafitte, Tomas Gajarsky and I, will be attending ISMIR 2019 in Delft (The Netherlands). This year, two of my colleagues from Moodagent will be presenting their work at ISMIR:

  1. Reinier will be presenting his paper, titled “JosquIntab: A Dataset for Content-based Computational Analysis of Music in Lute Tablature” in the main conference.
  2. Tomas will be presenting his paper, titled “Reinforcement Learning Recommender System for Modelling Listening Sessions” in the Late-breaking session of the conference.

Do stop by at these posters to learn more about these interesting topics!

Remote Talk at Event Organised by Music Tech Community – India

- lmn

I was invited by the Music Tech Community – India (MTC – India) to deliver a talk on the 29th of December, 2018 in Bengaluru. The theme of the event was “Machine Learning for Art & Music Generation” where my work at Jukedeck fit in perfectly alongside that of the other speakers at the event.

I happened to be on a holiday then in beautiful Mararikulam in Kerala around then, but I really didn’t want to miss this opportunity to speak so we decided to make it a remote talk that I delivered via Skype. Thanks to the excellent organisers – Albin Correya, Manaswi Mishra and Siddharth Bharadwaj, the talk went off smoothly and was apparently well-received. Other speakers during the event were Harshit Agarwal, and two of the organisers themselves – Albin Correya and Manaswi Mishra.

Oral Presentation at the 19th International Society for Music Information Retrieval Conference

- lmn

A few months following the acceptance of our paper at ISMIR 2018, I attended the conference in Paris with several of my colleagues from Jukedeck. We had a fairly large presence there dwarfed (as far as I can tell) only by a larger one from Spotify. The conference was organised very well and everything went-off smoothly. It was great to be back in the beautiful city after my last visit nearly 8 years ago!

I was particularly pleased by the new format for presenting accepted papers at this ISMIR wherein each paper was given both oral and poster presentation slots thus removing the traditional distinction between papers that exists in conferences. In the case of our paper on StructureNet, I made the oral presentation and my colleagues and co-authors – Gabriele and Marco – made the poster presentation. Fortunately, this year ISMIR was streamed live and the videos were later stored on YouTube so I’m able to share the video of my presentation with you. It’s only a 4-minute presentation so do check it out! And it appeared to me each time I passed our poster by that it received a lot of attention, and this was of course great! I, with help from members of my team, also prepared a blog post on StructureNet which was published recently on Jukedeck R & D Team’s Medium page. I urge you to give it a read if you’re curious what the paper is all about. Here’s a picture of the Jukedeck team at ISMIR:

The Jukedeck Team at ISMIR 2018 – (from left-to-right) Ben, Reinier, Gabriele, Matt, me, Katerina and Marco.

I also signed up to play in this year’s ISMIR jam session organised by Uri Nieto from Pandora! If I remember correctly, it’s something that started in 2014 and has been getting more popular by the year. As anticipated, the jam session was a success and a lot of fun, with music ranging from AI-composed folk tunes to Jazz, Blues, Rock and Heavy Metal. I played two songs with my fellow attendees – Blackest Eyes by Porcupine Tree and Plush by Stone Temple Pilots. My friend Juanjo shared a recording of the first song with me in which I played bass.

As always, ISMIR this year provided a great opportunity to make new acquaintances, and meet old friends and colleagues. As it turns out quite a few of my friends from the Music Informatics Research Group (MIRG) at City, University of London showed up this time and it was great to catch up with them.

The MIRG at ISMIR 2018: (from left-to-right, back-to-front) Shahar, me, Daniel, Tillman, Andreas, Radha and Reinier.

And to top it all off, my master thesis supervisor Hendrik Purwins managed to make it to the conference on the last day giving me the opportunity to get this one selfie with Tillman (my PhD thesis supervisor) and him.

Tillman, me and Hendrik at the conference venue.

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.

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.

Reflections on Three Months of Remote Work

- lmn

Beginnings…

It started with my wife Nina and I deciding that it would be good for us to move to Hyderabad for three months starting Oct 2, 2016 until Jan 8, 2017 for various reasons. As I was keen on continuing work at Jukedeck, I proposed the idea of me working remotely during this period to Patrick, the COO of the company. After some deliberation and another meeting with Ed (the CEO), much to my delight, the company decided to give it a try under the condition that we would review this arrangement each month and be quick to act in case of any unexpected (negative) eventualities. I was very excited and at the same time anxious as this was the first time I ever worked remotely from home.

Preparation

Before leaving London, I had a quick meeting with my team lead Kevin who was very supportive of this idea and we discussed a few things while leaving others to be dealt with as and when needed. First, we decided that I would be working from 11AM until 7PM IST instead of my usual working hours of 9:30AM to 5:30 PM BST which would (considering the 4.5 hours difference in time between Hyderabad and London) give me five hours of overlap in time with my team and three hours during which I would be by myself. We did also note that daylight savings time would set clocks in the UK one hour behind making the time difference 5.5 hours from the initial 4.5 hours but agreed to consider the option of me starting an hour later when this happens. We also agreed that I would update my team with my work every morning on the standup channel we have on Slack. If there were any brainstorming meetings, I would have the opportunity propose ideas before the meeting and again after going through the Google Doc containing the minutes of the meeting after it finishes. And I would be in touch with my team through Slack and, whenever needed, Skype. We also discussed a few worst case scenarios where, if this arrangement did not work out, I would consider switching to part-time work or even a sabbatical leave until my return to London in January.

Setting Things Up

On arrival in India, without any delay my first task was to setup an office at home. My parents, who we were living with during these three months, allowed me to use the guest room/study as my office. I had a reasonably quiet space with a big enough desk to work on. Although it went through a few changes as time went by, it essentially looked like this:

Home Office

Once I got started this way, I was ready to go! In the rest of this post, I’ll write about some of the things that stand out in my memory from others that were more mundane and easy to forget.

Participating in Standups Remotely

About two weeks into my move out of London I started noticing that I was unable to keep up with what some of my teammates were working on. I realised that this was due to the fact that while I was updating everyone with my work on Slack, the reverse was not happening. I brought this up with Kevin and we decided that the simplest thing to do would be for me to attend standups via Skype. We started first with one of the team members holding a laptop during standups with a Skype session which turned out to be a bit cumbersome in addition to the poor audio/video quality. Switching to a mobile phone was less cumbersome but still didn’t help the quality. We then came to know that it was possible to send video messages over Skype and while this was not real-time, it was certainly very clear and allowed me to go over standup in my own time. So we settled with this.

I suppose the bright side of this arrangement was that standups were brief, concise and to the point. There is a tendency for standups to turn into discussions about something very specific, involving only some of the team members while others wait without necessarily knowing what the conversation is about. It certainly avoided such a situation, and I even had a couple of my team mates acknowledge this benefit to me since we started with it.

Pair Programming Remotely

I was assigned a task at one point that required pair programming with my colleague, Marco. This was the first time for both of us to take part in remote pair programming. The first alternative we tried was to use the Atom editor plugin called atom-pair. It worked, however, as this was around the time when my broadband connection quality was at its worst the editor took several minutes to update the text that Marco typed, on my screen. It was bad. We then decided to switch to a more lightweight alternative as we have our trusty Jukedeck server, Ada. The setup was the following. We both connected to Ada via SSH. Once we were in, I started a Tmux session and opened the Python source file using Vim. Marco switched users to be me (he had superuser privileges on the server so I did not have to share my password with him for this) and attached himself to the same Tmux session from his end. Despite the lag, this worked like a charm! This setup came with the added benefit that we could open any number of shells through Tmux, and also have the IPython interpreter running alongside our editor to test our changes. While doing this, we also had a Skype session open where we discussed things. We carried on for about 5 hours with this with hardly any interruptions and got quite a bit of work done.

For the first attempt, I think this went very well. And a win for the very minimal command-line approach to work that I am strongly in favour of. As an alternative to both users involved in remote pair programming using one of the users’ accounts, a dedicated pair programming account can be created on the server which has access to all the relevant source repositories and to which multiple users have access. This would help if one or both of the users engaged in pair programming do not have superuser privileges.

Making Presentations Remotely

At Jukedeck, we have what we call Lunch & Learn (L & L) sessions where a member of the team (or someone the company invites) makes a presentation about a particular topic that might benefit or be interesting to others. I volunteered to do my first L & L session on “Machine Learning at Jukedeck” on Nov 1, 2016 where I planned to go over the basics of machine learning and how we employ it to power our AI music composer. The setup was fairly basic and simple. We initiated a Skype conversation on my colleague Eliza’s laptop, and I emailed a copy of my presentation to her so that she could navigate through it while I spoke from the other end. It went on smoothly without any interruptions and the message seemed to have gone across quite well. I did answer a few questions too, but couldn’t follow a few others due to a poor signal.

I found this to be a nice way to stay in touch with everyone else in the company (apart from my team members who I was liaising with everyday regarding work) and make my presence felt. I was keen on doing another L & L remotely, however, there was not enough time for this before my return to London.

Internet Issues

The only thing I wish had worked out better was my internet connection. Although we had a working 10 Mbps connection from BSNL (India’s National ISP), it was far from reliable. There were brief and frequent outages throughout the day on many days which was frustrating when loading webpages, pulling code changes from GitHub or working remotely via SSH on our company server. My only consolation was a patient and polite customer service, and the courteous technicians they sent forth to fix the connection. Fortunately, the worst of my connection woes lasted only during the first two weeks after which things got better.

To add to my troubles, the IP address (of my home router) from which I connected to our server in London kept changing on a daily basis, and since we had IP-based access restrictions in place I had to share my new IP every morning with Marco who would allow me to connect from it. We did this for about two weeks when we decided to simply unblock a range of IPs from which I seemed to be connecting. In contrast to this, in the UK, one’s public IP (say at home) does not seem to change over time which is what motivated this IP-based access restriction and made it possible in the first place. So now that I’m back, all those IPs are once again blocked and things are more secure once again.

Change in Working Hours

In the second month of my remote work, daylight savings time kicked in and I was one extra hour ahead of my team back in London. Kevin let me decide whether or not I would like to change when I started my day. Initially I did, so that I have the same number of overlapping hours with my team. After about two or so weeks, I found that this was not working out, mainly because I was almost completely losing the most productive part of my day – the morning hours before lunch. Plus, my day typically ended between 8-9PM and this nearly ruled out any prospects of making plans for the evening.

I decided to altogether break away from the daily 11-7 routine, and started even earlier in the morning on certain days when I didn’t anticipate much interaction with my team members. And by this time, both Kevin and I were seeing things work well and had the confidence that moving things around a bit is a minor risk to take if there was a chance of me being more productive. And it certainly didn’t make things worse!

Change of Location

It was around a month and a half after I first started working from my parents’ guest room when my cozy little home-office in the study stopped feeling as cozy. It felt isolating, and I just didn’t look forward to going in there every weekday morning. Clearly, I needed a change of environment. I’d been reading some books about remote work around that time (more on these below) and they suggested either trying out cafes or coworking spaces which would be bustling with some activity that might alleviate the feeling of isolation and lead to a healthier state-of-mind.

For a start, I moved to the dining room. This helped, as it was a bigger space and I’d see people more often than I did in the study. A friend of mine also put me in touch with one of his friends (a senior of mine from the IIIT-Hyderabad) told me of one of his batchmates who had gone on to found his own Data Science startup Predera that had an office in Hyderabad, and he was more than happy to let me work from there if I wanted to. As the office was at least a 30 minute drive away from where I was, I kept postponing my visit and ultimately didn’t end up going there but it was certainly very generous of him to keep the offer open!

The Ups

What I found particularly nice about this setup was that there were very few distractions and thus it was a joy to code, review research literature and GitHub pull requests thoroughly. Furthermore, any conversation I had with my colleagues related to work was concise and to the point. I was in the quiet comfort of my home and the nature of my work which mostly involved individual work with the occasional discussion with a colleague or two was well-suited for a remote setup. Personally, I didn’t find it hard to motivate myself to stick to a work schedule and I would like to attribute this to four years of learning to do this during my PhD. I felt more often like I did justice to the work I took up because of the lack of distractions. One can almost see this as an exercise in self-discipline.

I was also able to skip the roughly 1 hour long daily commute between my home and the office, which was a noticeable change. I wouldn’t really count this a positive change as I cycle to work everyday in London and rather enjoy doing it. However, I can imagine that for someone who drives or takes the public transport to work to work this might come as a change that they would welcome. In my case, I spent the hour I saved on exercise and running in a nearby park so there was not so much difference in what I spent the time for.

And I was also reminded on several occasions during the three months, what a fantastic team at Jukedeck I was a part of! All my team members were supportive of my move, patient and creative in dealing with any glitches that arose thereof and not once showed any signs of disapproval. Kevin was very good at assigning tasks to me that were both challenging and that I could work independently on with some discussion with others in the team. This really minimised delays and feelings of anxiety in me that I wasn’t able to contribute which could have ensued otherwise.

The Downs

While the experience was mostly positive, there were of course some downsides that became evident after just he first couple of weeks. As much as I did make the effort to share my input before and after the meetings, I felt a little less in control when it came to the direction in which the meetings went as this usually involved debating and persuasion which were much easier to do by being there in person. This effect would be far less pronounced if we were a fully remote company, in which case our processes would work the same for everyone. However, this wasn’t the case and while many things did work out, good communication was the biggest challenge among those that didn’t.

Physical absence from the office did feel isolating on a few occasions. I made up for this by engaging in the occasional friendly banter with colleagues over Slack and responding to their non-work related posts which helped me feel like I was still a part of what was happening in the office. It was also important to get out of the house a couple of times during the week just for some change of environment otherwise I felt locked-in. I missed all the team outings and lunches which was a very good opportunity to bond at a personal level with my teammates.

I did see myself falling behind some new developments taking place in the office, particularly those that came about during meetings that I was unable to attend due to the time-difference or poor communication between me and the team in London during the meeting. At least in my case, Skype (or Slack video chat) did not work as well as I had hoped they would. I would say the success-rate was around 40%.

Again, these downsides were not something that couldn’t be addressed but I thought it would only be fair to mention them along with the things that did work out. I’ll not speculate about how it would have turned out otherwise, but I sure was happy to be back in person to the Jukedeck office in London after three months.

If You’re Interested in Remote Work Too…

There is an excellent Hacker News post that answers several questions related to remote work, and also contains some very handy links to websites that facilitate remote work and create opportunities for those seeking to make a career out of working remotely. Here I came across two very well-written books on working remotely. The first is called “Remote: Office not Required” by the founders of 37Signals (now Basecamp) which is a company that has seemingly mastered the art of effective remote work. And the second is “The Ultimate Guide to Remote Work” by the folks at Zapier. You can even get a free PDF/MOBI copy of this book on their website. While there is some degree of (I should note, reasonable) self-marketing that went into both these books, they are definitely well worth reading for anyone wanting to get an insight into the pros and cons of remote work. Essentially, what all the different resources gathered in this Ask HN page suggest is that thanks to technology, we’re heading towards a world where remote work (at least in the tech sector) is becoming more and more feasible for those seeking a change from the 9-5 office work. It certainly gave me something to relate to, tips to follow and a feeling of being a part of a larger (but not large in an absolute sense) movement.

In Retrospect

As much as I had my apprehensions (as I often do with many things), I think this was a fantastic experience overall – I got to spend time with my parents after nearly two years of being away in the UK busy with my PhD, meet old friends, get married to my lovely wife Nina who I must thank for insisting on moving to India for three months, and last but obviously not the least be a part of a work arrangement that was indeed something new and unique in my experience. It got me interested and researching about making a career working remotely, which is something I believe I’m likely to follow up on at some point later in my career.

A Visit to MusicMuni Labs

- lmn

I was in Bangalore last weekend with my wife Nina who delivered a talk on Music Therapy for Dementia at ARDSICON 2016. During my stay there, I took the opportunity to meet a couple of friends at MusicMuni Labs, a startup that is working on some very cool apps for Music Education in India. This post talks a little about MusicMuni and what they’re up to.

To give you some background, this startup is the brainchild of two of my friends Gopala Koduri and Sankalp Gulati together with their mentor Prof. Xavier Serra and other music technology researchers at the Music Technology Group (MTG) of Universitat Pompeu Fabra. The MTG has created several successful startups in the past, and this is one of their newest ventures that employs the research that has been carried out as a part of the CompMusic project among other state-of-the-art music technology research for learning and exploring music of the Hindustani and Carnatic traditions.

The team has so far released two apps for Android – Riyaz and Sarāga, both currently in the beta stage and steadily gaining a user base. This is how the team describes Riyaz which is apparently their main focus at the moment:

“This android application aims to facilitate music learning for beginner to intermediate level music students by making their practice (riyaz) sessions more efficient. This application includes cutting edge music technologies that employ perceptually relevant models to automatically evaluate how well a student is singing compared to a reference music lesson. Students get a fine grained feedback on their singing.”

And Sarāga is described as follows:

“Sarāga is an android application that provides an enriched listening atmosphere over a collection of Carnatic and Hindustani music. It allows Indian art music connoisseurs and casual listeners to navigate, discover and listen to these music traditions using familiar, relevant and culturally grounded concepts. Sarāga includes inclusive designing of innovative visualizations and inter and intra-song navigation patterns that present musically rich information to the user on a limited screen estate such as mobiles. These time synchronized visualizations of musically relevant facets such as melodic patterns, samas locations and sections provides a user with better understanding and appreciation of these music traditions.”

They’re a very early stage startup with a very small and dedicated team, so I wish them all the very best and look forward to exciting updates from them in the future. Do check out their apps on the links I shared above in this blog if you interested in classical music of India. And if you’re looking to do an internship with a passion for music and music technology, they would be happy to hear from you!

MusicMuni Labs – Swapnil, Utkarsh, Gopala and Sankalp (from left-to-right).

 

Oral Presentation at the 28th International Joint Conference on Neural Networks

- lmn

My paper was accepted accepted for oral presentation at the 28th International Joint Conference on Neural Networks, held in the picturesque town of Killarney in Ireland. The title of the paper is quite a mouthful – “Discriminative Learning and Inference in the Recurrent Temporal RBM for Melody Modelling” and its abstract is the following:

“We are interested in modelling musical pitch sequences in melodies in the symbolic form. The task here is to learn a model to predict the probability distribution over the various possible values of pitch of the next note in a melody, given those leading up to it. For this task, we propose the Recurrent Temporal Discriminative Restricted Boltzmann Machine (RTDRBM). It is obtained by carrying out discriminative learning and inference as put forward in the Discriminative RBM (DRBM), in a temporal setting by incorporating the recurrent structure of the Recurrent Temporal RBM (RTRBM). The model is evaluated on the cross entropy of its predictions using a corpus containing 8 datasets of folk and chorale melodies, and compared with n-grams and other standard connectionist models. Results show that the RTDRBM has a better predictive performance than the rest of the models, and that the improvement is statistically significant.

I presented the paper in the session on Recurrent Neural Networks. The model that we proposed in the paper – the RTDRBM – was the first original Machine Learning contribution of my PhD. And it was a pleasure to collaborate with my friend and colleague Son Tran in the work. He presented a second paper at the conference titled, “Efficient Representation Ranking for Transfer Learning” .

With Son and my supervisor Artur after my presentation.

Yet again a conference has taken me to a place in the world that I probably would’ve never visited otherwise! This doesn’t at all mean that the visit wasn’t worthwhile. The lush green Irish landscape, the charming town of Killarney and the abounding nature around it, and a friendly and welcoming hostel all made this a very memorable trip! Unfortunately, I had sore throat and a fever during much of my stay so I chose Irish coffee over a pint of Guinness (which I heard tastes much better in Ireland) when I had the chance. I regret this, but maybe that’s another reason to visit Ireland once again sometime!

On one of my healthier days in Killarney.

Oral Presentation at the 15th International Society for Music Information Retrieval Conference

- lmn

We had two papers accepted at the 15th International Society for Music Information Retrieval Conference (ISMIR). Given the fantastic experience I had at ISMIR the year before, I was super-excited to travel to Taipei to attend the conference. The first of these papers is titled, “Multiple Viewpoint Melodic Prediction with Fixed-Context Neural Networks” and is in some ways a continuation of my work from the previous ISMIR conference. The abstract of the paper is as follows:

“The multiple viewpoints representation is an event-based representation of symbolic music data which offers a means for the analysis and generation of notated music. Previous work using this representation has predominantly relied on n-gram and variable order Markov models for music sequence modelling. Recently the efficacy of a class of distributed models, namely restricted Boltzmann machines, was demonstrated for this purpose. In this paper, we demonstrate the use of two neural network models which use fixed-length sequences of various viewpoint types as input to predict the pitch of the next note in the sequence. The predictive performance of each of these models is comparable to that of models previously evaluated on the same task. We then combine the predictions of individual models using an entropy-weighted combination scheme to improve the overall prediction performance, and compare this with the predictions of a single equivalent model which takes as input all the viewpoint types of each of the individual models in the combination.”

The paper was presented as a poster. The second paper is based on very interesting work I did in collaboration with Siddharth Sigtia and Emmanouil Benetos on automatic transcription of polyphonic music, titled “An RNN-based Music Language Model for Improving Automatic Music Transcription” that Siddharth presented as another poster.

I have to note that this year’s ISMIR organisation was fantastic! Everything from the review process, information on the website to the venue, the assitance at the venue, and the banquet were very well managed and executed by the organisers. The most interesting part of the conference for me was the keynote lecture, titled “Sound and Music Computing for Exercise and (Re-)habilitation” by Prof. Ye Wang, in which he described the potential in music to serve as a means to rehabilitate and improve the quality of life of individuals with different ailments, and illustrated this with the help of a few projects his group at the National University of Singapore has been working. It was a very inspiring talk, and I really admire Dr. Wang’s statement regarding the often overlooked direct impact of research and published work to society which has been the cornerstone of these projects. I have lately taken interest in Music Therapy and have been going through some literature to see if my own work on music modelling can in some way be applied to achieve therapeutic goals. There were some interesting late-breaking sessions as well that I took part in, including the very successful one organised by my supervisor Tillman on Big Data and Music where I was taking notes during the discussion.

And finally, as is always the case when I attend a conference, I did take some time off in Taipei and its surrounding areas. On one evening, I joined some friends and colleagues to go see the tallest building in the city – Taipei 101.

Jan and I with Taipei 101 in the background (Photo Courtesy: Marius Miron)
Jan and I with Taipei 101 in the background (Photo Courtesy: Marius Miron)

On another day, a couple of us planned a day-trip to a nearby village called Jiufen where we checked out some temples, the market and the old Japanese mining village on top of a hill.

The gang that went on a day-trip to Jiufen, led by our lovely host Kailie (first from the right).
The gang that went on a day-trip to Jiufen, led by our lovely host Kailie (first from the right).

And on another day, I joined my buddy Marius on a local site-seeing round to see some local museums, Shilin night market, Chiang Kai Shek Memorial, and other places before taking the long flight back to London eventually.

Taipei was fantastic, and I’d be up for another visit anytime! Last but not least, the hospitality of Fun Taipei hostel made the whole trip a little better each day.

Poster Presentation at the Machine Learning Summer School

- lmn

I was selected to attend the Machine Learning Summer School in Reykjavik between April 25-May 4, 2014. I was also awarded a travel grant to attend this event which made it possible for me to attend it. I also proposed to present a poster about my ongoing work on musical pitch prediction with neural networks.

Many of the topics were very new to me, but I found the tutorials on Machine Learning and HCI (Roderick Murray-Smith), Introduction to ML (Neil Lawrence), Deep Learning (Yoshua Bengio), Probabilistic Modelling (Iain Murray), and Reinforcement Learning (David Silver) particularly interesting. Especially the last talk seemed like there was much in it that could be adopted into my own work on music modelling and I was very tempted to do so. Let’s see how that goes.

I was also a bit stressed carrying out experiments for a paper we’re submitting to the 15th International Society of Music Information Retrieval Conference (ISMIR 2014). So fingers-crossed that it will all work out for that.

I managed to travel a little while I was in Reykjavik. This was something that had to be done given how novel a destination Iceland is. I joined the rest of the workshop attendees on the Golden Circle Tour that showed us some fascinating and very alien Icelandic landscapes.

Mount Esjan as seen from my hostel (Kex Hostel) in Reykjavik. Iceland is full of stunning nature!

And finally, I made a last-minute trip to the Blue Lagoon on the day before my return to London.

Last-minute decision to visit the Blue Lagoon that certainly paid off!

It was indeed very fortunate that I was able to attend the summer school in Reykjavik. This has been an incredible learning experience one of the most unique destinations I have been to in my entire life!

I’m sharing a copy of the poster (made using Beamer/LaTeX) I presented here.

MLSS 2014 Presentation