Farhi was trained as a theoretical particle physicist but has also worked on astrophysics, general relativity, and the foundations of quantum mechanics. His present interest is the theory of quantum computation. As a graduate student, Farhi invented the jet variable "Thrust" which is used today at the Large Hadron Collider to describe how particles in high energy accelerator collisions come out in collimated streams. He then worked with Leonard Susskind on grand unified theories with electro-weak dynamical symmetry breaking. At CERN, he and Larry Abbott proposed an model in which quarks, leptons, and massive gauge bosons are composite. At MIT, with Robert Jaffe, he worked out many of the properties of a possibly stable super dense form of matter called ``Strange Matter" and with Charles Alcock and Angela Olinto he studied the properties of ``Strange Stars", compact objects made of strange matter. His interest then shifted to general relativity and he and Alan Guth studied the classical and quantum prospects of creating a new inflationary universe in a laboratory today. He and Guth, along with Sean Carroll, showed how building a time machine would require resources beyond what could ever be possible to obtain. Since the late '1990s, Farhi has been studying how to use quantum mechanics to gain algorithmic speedup in solving problems that are difficult for conventional computers. He and Sam Gutmann pioneered the continuous time Hamiltonian based approach to quantum computation which is an alternative to the conventional gate model. He and Gutmann then proposed the idea of designing algorithms based on quantum walks, which was used to demonstrate the power of quantum computation over classical. They, along with Jeffrey Goldstone and Michael Sipser, introduced the idea of quantum computation by adiabatic evolution which generated much interest in the quantum computing community. For example, the D-Wave machine is designed to run the quantum adiabatic algorithm. In 2007, Farhi, Goldstone and Gutmann showed, using quantum walks, that a quantum computer can determine who wins a game faster than a classical computer. In 2010, he along with Peter Shor and others at MIT introduced a scheme for Quantum Money which so far has resisted attack. In 2014 Farhi, Goldstone and Gutmann introduced the Quantum Approximate Optimization Algorithm, a novel quantum algorithm for finding approximate solutions to combinatorial search problems. The QAOA is viewed as one of the best candidates to run on Noisy Intermediate Scale Quantum devices which are coming online in the near future. Farhi continues to work on quantum computing but keeps a close eye on particle physics and recent developments in cosmology.
