QIP 2014

I have around 5 hours to spare in Frankfurt coming back from Barcelona on my way back home to Toronto. I was in Barcelona for QIP 2014: a fantastic conference in an even more extraordinary city. The entire conference experience was enriching and stimulating, so I thought I could spend my time in Germany writing about it.


Conference Group Picture

I heard a few people mentioning that this was the largest instalment of QIP, which seems reasonable when you take a look at the conference picture. I like to think that this is due not only to the convenient and attractive location, but because our field keeps attracting more people and continues to have an influence in increasingly more areas. Daniel Gottesman told me over dinner that when he started working in quantum information, there were conferences organized where basically the entire field would attend, both theorists and experimentalists. Together they would encompass around 20 people. The attendees of QIP 2014 were estimated to be around 400. It is exciting to wonder just how the field will look like in a couple of decades.

Barcelona is an ideal place to host a medium-sized conference like this one. The venue for the talks (AXA Auditorium), the rump session (Centre de Cultura Contemporània de Barcelona) and the conference dinner (Museu Marítim de Barcelona) were all plain gorgeous. I guess this is not hard to achieve in a city where basically every other building is a standing work of art; not something you can say about Waterloo, for example. You could also feel the desire of everyone to explore the city during lunch and after each day’s session. This, combined with the apparent Catalan tradition of taking hours to eat (seriously, it was nearly impossible to have dinner in under 3 hours), was a great formula to have attendees spend time together discussing and exchanging ideas. I most definitely benefited from meeting many people and having very interesting conversations.

Barcelona as seen from atop Parc Güell.

Barcelona as seen from atop Parc Güell.

There were many great talks on a wide range of subjects: very impressive progress keeps being made in the field. Initially, I had expected the content of the talks to be heavily biased towards subjects in computer science and mathematics, and I was concerned that perhaps I would have trouble understanding or being interested in them. I’m glad to say that my experience was quite the contrary: I found myself being able to follow most of the talks and was surprised by how fascinated I was with topics I hadn’t previously encountered in enough detail. I imagine this is a consequence of the exposure I get to the mathematical aspects of quantum information as a student in the Institute for Quantum Computing.  Having said that, I would like to offer a ranking of my personally favourite talks, which is heavily subjective as I am counting mostly how compelling I considered them to be.

5. Purifications of multipartite states: limitations and constructive methods. Speaker: Gemma de las Cuevas.

A very clear exposure of the problem of finding efficient tensor-network representations of multipartite mixed states. It made me think that a solid knowledge of tensor networks may be a powerful tool to add to the arsenal of a quantum information theorist. This is especially true since, according to the speaker, they may find applications in quantum communication complexity.

4. Robust protocols for securely expanding randomness and distributing keys using untrusted quantum devices.  Speaker: Carl Miller.

Even though I disliked how much advertising was involved in this talk (he was already referring to his own protocol as the Miller-Shi protocol!), the results presented are impressive. I have found randomness to be interesting for as long as I can remember, and the prospect of using fundamental aspects of quantum mechanics to generate reliable random numbers is an exciting one for me. The work presented has gone a long way in elucidating how an ideal random number generator may be constructed.

3. What is the overhead required for fault-tolerant quantum computation? Speaker: Daniel Gottesman.

The essential message of Daniel’s talk was this: we should not be satisfied with the overhead required in current fault-tolerant schemes. As he eloquently stated, we may dream of a quantum computer with one thousand physical qubits, but we would not have gained much if we can only operate on ten logical qubits. To back up his message, he provided a theorem that guarantees that the overhead can be made arbitrarily small in principle. *Restrictions may apply.

2. Classification of the complexity of local Hamiltonian problems. Speaker: Ashley Montanaro.

There had been previously a lot of interest in determining the complexity of several versions of the local Hamiltonian problem. These guys classified an enormous class of problems in one single paper. Amazing. Bonus awarded for using the face of Toby Cubitt to denote a qubit (how many Cubitt-Qubit puns must he have experienced?).

1. Universal fault-tolerant quantum computation with only transversal gates and error correction. Speaker: Adam Paetznick.

I am obviously biased since Adam is a good friend of mine, but I love the result he presented. My favourite kinds of breakthroughs are those that look incredibly simple in hindsight, but that no one had been able to come up with before. Transversality is a highly desired property of the implementation of fault-tolerance schemes for quantum computation, but there was a long-held belief that this was impossible due to a no-go theorem by Eastin and Knill. Adam and his supervisor Ben Reichardt circumvented this fake barrier by simply adding the already existing ingredient of error correction into the mix. His explanation of the basic idea was brilliantly done, and I am sure he had many people in the audience wondering why they hadn’t thought of that before.

I didn’t include any plenary talks into my ranking because I consider them to be closer to a colloquium than to the presentation of new results. Nevertheless, I enjoyed plenary talks greatly, and my favourite was the talk by Matthias Troyer on the classical simulation of complex quantum systems, largely because of how much I learnt. He did an amazing job at showcasing the competition that future quantum computers must face: extremely fast classical supercomputers. As he put it, “it is hard to build a quantum machine that can beat my codes.” Using Quantum Monte Carlo Methods, we are currently able to simulate the equilibrium properties of millions of bosons, a tough task for a quantum computer to surpass. My understanding is that only systems of interacting fermions with a sign problem are hard to simulate classically, so perhaps that’s where quantum computers can really shine. However, Matthias also pointed out that something classical simulators cannot do well is to simulate dynamics: “If I kick a quantum system, no one knows how to simulate what happens.”

He also included a list of the most cited papers in physics, with the vast majority being papers that described methods to solve a particular problem or carry out a simulation. This leads me to the conjecture that the best way to produce a highly cited paper is to find a tool that a large community of working scientists will use.

The poster sessions were, unfortunately, an extremely ineffective way of showcasing research.  The two sessions included around a hundred posters each, in a small crowded room where it was nearly impossible to hear someone who was further than 30cms from your face. During the presentation of my own poster, I had many people stop over, which was great, but literally prevented me from even glimpsing half of the posters presented. If I had to pick my favourite poster, I would probably choose one by Valerio Scarani and his students on the tree complexity of quantum states, but only because it was very relevant to my research in quantum communication complexity.

As this was my first QIP, it was also my first rump session. The evening was a combination of hilariousness, surrealism and schadenfreude. The highlights of the night were the appearance of the Schroedinger’s rat, the Zurich toast protocol (which was largely implemented during the conference dinner) and John Smolin’s talk demolishing D-Wave. I am proud to confess that I was the one who ironically shouted “Go D-Wave!” during the beginning of the talk. Hearing him say “Did someone seriously just say ‘Go D-Wave!’” was priceless. I also laughed intensely when, asking for champagne (in Spanish) to one of the waiters, he took the opportunity to ask me: “Hey, just what the hell is going on here?” I guess that from an external perspective, it all must have seemed incredibly bizarre. Finally, my advice to anyone planning to speak at future rump sessions: Do not talk about your research in any level of seriousness! You will only achieve people ignoring you in a very painful manner.

John Smolin during the Rump Session.

John Smolin during the Rump Session.

On a different note, I was happy to see how good the community looked. Barring a few exceptions, we were a well-dressed, classy-looking group of people, largely destroying the prejudice of scientists as beings who wear t-shirts, non-matching pants and ugly sneakers. I believe we should look as smartly as we actually are.

Sadly, the conference provided a great way of estimating the number of women doing research in quantum information, with the numbers being embarrassingly small. Out of 40 speakers, only three were women (7.5%) and out of the 23 members of the programme committee, only two were women (8.6%). These numbers roughly match what one could estimate by looking at the audience as well. In my opinion, it is worrisome that such a small percentage of the members of our community are women and I truly hope this figure will change significantly in the near future.

A conference like QIP is a great place to put your research into perspective. It is very easy to get lost in obsession with a particular problem and easy to lose grasp of what the rest of the world is worrying about. Part of being a good researcher is asking the correct questions, and this is easier to do when you are aware of all the questions everyone else is asking. Additionally, it allows you to realize how much your own results may impact the community. Overall, QIP 2014 was a wonderful experience and I am already looking forward to the possibility of being in Sydney for QIP 2015.

I love the sea.

I love the sea.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s