I recently read Seth Lloyd's A Turing Test for Free Will -- conveniently related to the subject of the blog's last piece, and absolutely engrossing. It's short, yet it makes a wonderful nuance in the debate over determinism, arguing that predictable functions can still have unpredictable outcomes, known as "free will functions."
I had thought that the world only needed more funding, organized effort, and goodwill to solve its biggest threats concerning all of humanity, from molecular interactions in fatal diseases to accessible, accurate weather prediction for farmers. But therein lies the rub: to be able to tackle large-scale problems, we must be able to analyze all the data points associated to find meaningful recourses in our efforts. Call it Silicon Valley marketing, but data analysis is important, and fast ways of understanding that data could be the key to faster solution implementation.
Classical computers can't solve almost all of these complex problems in a reasonable amount of time -- the time it takes for algorithms to finish increases exponentially with the size of the dataset, and approximations can run amok.
Quantum computers, using qubits instead of bits as the basic units of data storage, could theoretically cut the time needed to process algorithms by leaps and bounds. A problem that could take years could be reduced to seconds, and policymakers can more quickly integrate data into their proposals, figuring out more efficiently how to allocate resources or support community welfare. My idealism powers my studies in quantum computing.
But idealism eventually has to run into reality, and the truth can be disheartening. We are years away from any semblance of commercially viable quantum computers, much less those that can readily implement machine learning and neural networks. Heavy skepticism, albeit necessary for scientific rigor, accompanies every new announced breakthrough, and the most interesting ideas are still elusive beyond special laboratory environments. I was thrilled to read about how a team at Harvard successfully modeled how proteins fold, the problem I'm currently working on, using a quantum computer from D-Wave, only to be dismayed when they got 13 out of 10,000 cases correct. I want solutions to address urgent global challenges immediately, but I forget that quantum computing is a nascent field, where theory far outpaces experimental implementation.
The Paris climate talks ended on a positive note, yet I couldn't shake the sense that the status quo still remained, undisturbed by this monumental announcement. Just because the world is making improvements doesn't mean it is free of problems; limitations in computing power still limit the potential for accelerated drug discovery for spreading tropical diseases, and island nations are still at risk of being swallowed up by rising seas. But my initial helplessness at the vastness of these problems has now only sharpened my purpose. I tear through initially unfamiliar papers on Quantum Key Distributions, a potential method of encryption using quantum computing, and quantum algorithms, banking on exponentially increased computing power to tackle the most currently intractable problems, from molecular simulations to climate change models to stock market prediction for systemic risk.
My choice of field in physics partially reflects my belief in quantum computing to, at some point, achieve significant gains over classical computing in a significantly wide variety of problems. This is a future I may see either only decades from now or, perhaps, never. But the problems in quantum computing are exciting, fun, and I get the sense I can be creating the next leap in technology, however overhyped that phrase is. I certainly wish to soon see the day the first viable quantum computer is demoed, but I may not, and I must come to terms with that. That didn't stop any of the previous generations of inventors, and that won't stop me.