Quantum-Enhanced Machine Learning

The emerging field of quantum machine learning has the potential to substantially aid in the problems and scope of artificial intelligence. This is only enhanced by recent successes in the field of classical machine learning. In this work we propose an approach for the systematic treatment of machine learning, from the perspective of quantum information. Our approach is general and covers all three main branches of machine learning: supervised, unsupervised, and reinforcement learning. While quantum improvements in supervised and unsupervised learning have been reported, reinforcement learning has received much less attention. Within our approach, we tackle the problem of quantum enhancements in reinforcement learning as well, and propose a systematic scheme for providing improvements. As an example, we show that quadratic improvements in learning efficiency, and exponential improvements in performance over limited time periods, can be obtained for a broad class of learning problems.

Figure 1

Tested agent-environment interaction. In general, each map of the tester $U_k^T$ acts on a fresh subsystem of the register $R_T$, which is not under the control of the agent, nor of the environment. The crossed wires represent multiple systems.

Figure 2

Differences between the interaction for $A$ and $A^q$. In steps 1 and 2, $A^q$ uses access to $E_{\text{oracle}}^q$, for $O(t)$ steps, and obtains a rewarding sequence $h_r$. Step 3: $A^q$ simulates the agent $A$, and trains the simulation to produce the rewarding sequence. In step 4, $A^q$ uses $A(h_r)$ for the remainder of the now classically tested interaction, with the classical environment $E$.