Quantum Internet Search Application

Seth Lloyd is proposing the first quantum app, or q-app, which he calls ‘Quantum Machine Learning for Big Quantum Data

Seth Lloyd’s q-app (pronounced “quapp”) encodes Google-like queries with q-bits that enable quantum computers to not only perform real-time searches through even the most gigantic databases, but which also insures their absolute privacy, since attempts to eavesdrop on the query by the search engine provider would disturb the delicate q-bit’s superposition of states

Lloyd has tried to get commercial funding to develop his q-app, but has so far failed to convince any venture capitalist to fund his project. The reason, he claims, is that his q-app insures that the search engine would not be able to store the user’s queries to add to the reams of information they already store about each of their users.

“The VCs say that the search-engine business model is to learn everything they can about their users, so my q-app goes against the very core of their business,” said Lloyd.

To prove that it will work, and hopefully attract some brave VC to fund its development, Lloyd is revealing the details of how his q-app makes real-time searches through the biggest conceivable databases. And after losing control of the technology that he claims to have invented, which he says D-Wave is currently using without paying him royalties, this time he has patented his Quantum Machine Learning q-app.

Nextbigfuture has an article about Seth Lloyds technical papers on his Quantum machine learning.

Seth proposes uses the same types of machine learning algorithms that have already been developed for dealing with big-data, but encodes their queries with q-bits which can only be processed on quantum computers. He claims the technique is so powerful that it could search for results in real-time through the biggest possible data sets. For instance, if each person on the earth’s genome was sequenced and encoded on a vector, then an array of these vectors could represent the entire combined genome of every person on the earth, which would be about 10 to the 20th bits (1,000,000,000,000,000,000,000). However, Lloyd claims that even a gigantic database could be queried in real-time using his q-app running on a future quantum computer with only a 70 q-bit processor.

The key to Lloyd’s q-app is that it would use quantum mechanics to map these big arrays of big vectors onto a tensor product space that results in an exponential compression of the data set.

“We get a kind of quantum compression of the classical bits into much smaller q-bit space, then use conventional machine learning algorithms, but on the much smaller q-bit space,” said Lloyd. “Many of the most popular machine learning techniques would work in this much smaller quantum space.”

Lloyd has tested his algorithm — in theory and on a small scale — using a conventional supervised machine learning algorithm operating on his q-bit space that shrinks classical data sets exponentially.

Putting quantum machine learning into practice will be more difficult. Lloyd estimates that a dozen qubits would be needed for a small-scale demonstration.

Quantum self analysis (Arxiv – 7 page pdf)

The usual way to reveal properties of an unknown quantum state, given many copies of a system in that state, is to perform measurements of different observables and to analyze the measurement results statistically. Here we show that the unknown quantum state can play an active role in its own analysis. In particular, given multiple copies of a quantum system with density matrix R, then it is possible to perform the unitary transformation e^{-i R t}. As a result, one can create quantum coherence among different copies of the system to perform `quantum self analysis,’ revealing the eigenvectors and eigenvalues of the unknown state in time exponentially faster than any existing algorithm.

Quantum support vector machine for big feature and big data classification (Arxiv – 5 page pdf)

Supervised machine learning is the classification of new data based on already classified training examples. In this work, we show that the support vector machine, an optimized linear and non-linear binary classifier, can be implemented on a quantum computer, with exponential speedups in the size of the vectors and the number of training examples. At the core of the algorithm is a non-sparse matrix simulation technique to efficiently perform a principal component analysis and matrix inversion of the training data kernel matrix. We thus provide an example of a quantum big feature and big data algorithm and pave the way for future developments at the intersection of quantum computing and machine learning.

Quantum algorithms for supervised and unsupervised machine learning (Arxiv – 11 pages)

Machine-learning tasks frequently involve problems of manipulating and classifying large numbers of vectors in high-dimensional spaces. Classical algorithms for solving such problems typically take time polynomial in the number of vectors and the dimension of the space. Quantum computers are good at manipulating high-dimensional vectors in large tensor product spaces. This paper provides supervised and unsupervised quantum machine learning algorithms for cluster assignment and cluster finding. Quantum machine learning can take time logarithmic in both the number of vectors and their dimension, an exponential speed-up over classical algorithms.

If you liked this article, please give it a quick review on ycombinator or StumbleUpon. Thanks