A different white-box deep architecture? Is it possible?

I recently published a paper more thoroughly explaining what I have been developing as a part of my PhD research at the Laboratory for Computer Graphics and Multimedia, Faculty of Computer and Information Science.

The paper describes an alternative deep architecture. Those of you who are involved in recent machine learning research probably know neural-network-based deep architectures, which recently gained a lot of attention from general public due to their results in a variety of fields. However, they are not perfect (at least I think so … :) ). To put it in a simple description, these networks lack transparence. We know how they work, but cannot see what they learn and thus often use them as black boxes (provide input, get output, don’t care about the process). We propose a different kind of deep architecture based on compositionality (large complex things are built of small simple things).

We tested our model on music information retrieval tasks and are currently seeking problems in other scientific fields dealing with big data – please message me for suggestions or potential collaboration.

If you are interested, I welcome you to read our paper, (freely) available here: http://dx.doi.org/10.1371/journal.pone.0169411

For a quick overview, I attached the paper’s abstract below:

Robust Real-Time Music Transcription with a Compositional Hierarchical Model

The paper presents a new compositional hierarchical model for robust music transcription. Its main features are unsupervised learning of a hierarchical representation of input data, transparency, which enables insights into the learned representation, as well as robustness and speed which make it suitable for real-world and real-time use.

The model consists of multiple layers, each composed of a number of parts. The hierarchical nature of the model corresponds well to hierarchical structures in music. The parts in lower layers correspond to low-level concepts (e.g. tone partials), while the parts in higher layers combine lower-level representations into more complex concepts (tones, chords). The layers are learned in an unsupervised manner from music signals. Parts in each layer are compositions of parts from previous layers based on statistical co-occurrences as the driving force of the learning process.

In the paper, we present the model’s structure and compare it to other hierarchical approaches in the field of music information retrieval. We evaluate the model’s performance for the multiple fundamental frequency estimation. Finally, we elaborate on extensions of the model towards other music information retrieval tasks.

Full paper is available here: http://dx.doi.org/10.1371/journal.pone.0169411

Fixing Hammond XB-2

A few weeks ago Dejan and I began our work on Hammond XB-2 again. The current problem can be heard here:

We tracked the problem back to the static RAMs, which was also mentioned as a solution in one of the comments on this similar youtube video.

We opened the keyboard and took out the main board. As you can see in the picture below, somebody already played (changed) the MUSE RAM chips (the size is different, thus the chips are hanging by the wires – top right corner).

The XB-2 main board

The XB-2 main board

Suprisingly, this thing worked in its own distorted way …

When you change the chips, make sure that you have the same form factor

When you change the chips, make sure that you have the same form factor

We unsoldered the old chips and cleaned the board …

Cleaning the holes ...

Cleaning the holes …

We also made sure we have the right DIL (dual in-line) sockets and put those on first.

Lots of pins to solder ...

Lots of pins to solder …

The DILs

The DILs

Dejan obtained the replacement chips from old computer boards. These are not identical to the originals (the new ones must be newer since the old ones had 70ms latency according to the manual) but should do the job.

Chips in sockets

Chips in sockets


However, the problem remained. Next step: change the V61C16 chips. Unfortunately we don’t have those and need to find them (or their alternatives) first. If anyone has any information on where to get them, please let me know!