So the news already broke that I was in Beijing to attend this year’s QIP…and now I’m trying to fight jet-lag by writing a half-coherent post about the first day of the conference. Let’s see if it works out.
Of course today’s highlight was the glimpse of blue sky that we got at lunchtime (and no, you will not get to see the picture I took); not something that we expected from the previous day’s heavy fog. And even that was supposed to be much better than last week, during which record highs levels of PM2.5 were recorded (apparently, comparable to exhaling directly from the exhaust pipe of a 1000MW coal-based power plant). If you believe this piece though such levels were not uncommon in large industrial US cities not so long ago, before the health issues associated with pollution became obvious. According to Andy Yao’s opening speech the situation is improving even in Beijing; I certainly wish it is the case for the local population.
We’re hosted in the huge Mong Man Wai Concert Hall, itself at the heart of Tsinghua University’s grandiose campus (itself unfortunately in a less appealing suburb of Beijing — somewhat fancy but consequently a little boring). The hall is huge and pretty packed; I don’t know how many people are in attendance but it definitely feels like a lot, including a large number of enthusiastic students from the recently created Centre for Quantum Information (oh! that website I can access, and fast!). The first poster session, held this evening, had 75 posters and was absolutely crowded, almost too much; it seems like the architects planned a lot of outside space in these buildings (huge esplanades), but less so inside. The high attendance was justified though, with many excellent posters as always at QIP.
This morning’s session had some very nice talks about quantum multiplayer games, including a great plenary talk by Ben Reichardt on the problem of testing a quantum mechanical system by classical means. From my perspective the highlights of the day came in the last session (well, the parts of it that I managed to stay awake in), dedicated to what seems (at least to the ignorant observer) to be a growing trend in the application of quantum information to the study of thermodynamics. Although thermodynamics has traditionally been concerned with “the average behavior of very large numbers of microscopic constituents”, and accordingly studied using tools from statistical mechanics, its main object, the connection between energy and work, can really be studied at all scales. Already at last year’s QIP in Montreal there was a very nice talk by Sandy Popescu on “The smallest possible thermal machines and the foundations of thermodynamics” — I think the answer was a qutrit, or a quantum system with three levels; quite a small engine indeed.
Today Jonathan Oppenheim talked about work with Michal Horodecki on “Fundamental limitations for quantum and nano thermodynamics“; he argued that thermodynamical transitions at the nano scale are not always irreversible and gave ways to quantify that irreversibility. Then Michael Ben-Or gave a talk on work with Daniel Gottesman and Avinatan Hassidim on their paper “Quantum refrigerator“. Michael started by linking computation with energy: computation would be effortless but for the presence of noise, as it is the correction of errors which by erasing information (setting all those noisy bits to 0) requires energy. Their results examine how long one can push a quantum computation without making any effort, i.e. without introducing any fresh ancillas in order to provide for error correction. In the case of the most standard form of noise, the single-qubit depolarizing channel, the answer is not much: circuits are restricted to having depth logarithmic in the number of initial qubits. Their main result is that for other types of channels, such as the dephasing channel, one can go for much longer (of course the dephasing channel doesn’t affect classical computations), and they give a comprehensive classification of channels with respect to how long one can compute, without fresh ancillas, under single-qubit noise modeled by the corresponding channel.
PS: The title of this post is not meant to suggest that I will write anything in the following days — indeed I do hope to find better ways to fight jet-lag 🙂