Empirical Analysis of Decision Making of an AI Agent on IBM’s 5Q Quantum Computer

Abstract

A recent work has shown that using an ion trap quantum processor can speed up the decision making of a reinforcement learning agent. Its quantum advantage is observed when the external environment changes, and then agent needs to relearn again. One character of this quantum hardware system discovered in this study is that it tends to overestimate the values used to determine the actions the agent will take. IBM's five qubit superconducting quantum processor is a popular quantum platform. The aims of our study are twofold. First we want to identify the hardware characteristic features of IBM's 5Q quantum computer when running this learning agent, compared with the ion trap processor. Second, through careful analysis, we observe that the quantum circuit employed in the ion trap processor for this agent could be simplified. Furthermore, when tested on IBM's 5Q quantum processor, our simplified circuit demonstrates its enhanced performance over the original circuit on one of the hard learning tasks investigated in the previous work. We also use IBM's quantum simulator when a good baseline is needed to compare the performances. As more and more quantum hardware devices are moving out of the laboratory and becoming generally available to public use, our work emphasizes the fact that the features and constraints of the quantum hardware could take a toll on the performance of quantum algorithms.

Comparison of Two Quantum Nearest Neighbor Classifiers on IBM’s Quantum Simulator

“…In [3], we create some artificial datasets to visualize the working of a distance-based quantum classifier and extend their quantum circuit from binary classification to a multiclass classification. In [4], the training of an AI agent for its decision making is compared on an ion trap quantum system and on a superconducting quantum system, and discover that latter is more accurate than the former and tends to underestimate the values for the agent to make a decision when compared with the ion trap system. In [5], a quantum neuron is created with a nonlinear activation function like a sigmoid function If we increase the value of K, the prediction outcome may change.…”

“…Following the same principle as in [3][4][5], this current study chooses two related quantum nearest neighbor algorithms [2,6] and uses a simple dataset to demonstrate how they work, reveal their quantum nature, and compare their performances in detail using IBM's quantum simulator [7].…”

“…To better show the different classes appearing weighted by their distance to the test point, 4 ψ is rewritten as ( ) ( ) …”

“…Along in this direction, we have finished a few papers for this end [3][4][5]. In [3], we create some artificial datasets to visualize the working of a distance-based quantum classifier and extend their quantum circuit from binary classification to a multiclass classification.…”

Towards a Real Quantum Neuron

“…We have finished two reports recently using IBM's 5Q quantum computer [1,2]. One is analyzing a distance-based quantum classifier where we show the prediction probability distributions of this classifier on several well-designed datasets to reveal the inner working of this classifier and extend the original binary classifier quantum circuit to a multi-class classifier.…”