Moritz August

Moritz August, Mari Carmen Bañuls, Thomas Huckle

We present a method to approximate functionals $\text{Tr} \, f(A)$ of very high-dimensional hermitian matrices $A$ represented as Matrix Product Operators (MPOs). Our method is based on a reformulation of a block Lanczos algorithm in tensor network format. We state main properties of the method and show how to adapt the basic Lanczos algorithm to the tensor network formalism to allow for high-dimensional computations. Additionally, we give an analysis of the complexity of our method and provide numerical evidence that it yields good approximations of the entropy of density matrices represented by MPOs while being robust against truncations.

Moritz August, Thomas Huckle

We recently introduced a method to approximate functions of Hermitian Matrix Product Operators or Tensor Trains that are of the form $\mathsf{Tr} f(A)$. Functions of this type occur in several applications, most notably in quantum physics. In this work we aim at extending the theoretical understanding of our method by showing several properties of our algorithm that can be used to detect and correct errors in its results. Most importantly, we show that there exists a more computationally efficient version of our algorithm for certain inputs. To illustrate the usefulness of our finding, we prove that several classes of spin Hamiltonians in quantum physics fall into this input category. We finally support our findings with numerical results obtained for an example from quantum physics.

Moritz August, José Miguel Hernández-Lobato

In this work we introduce the application of black-box quantum control as an interesting rein- forcement learning problem to the machine learning community. We analyze the structure of the reinforcement learning problems arising in quantum physics and argue that agents parameterized by long short-term memory (LSTM) networks trained via stochastic policy gradients yield a general method to solving them. In this context we introduce a variant of the proximal policy optimization (PPO) algorithm called the memory proximal policy optimization (MPPO) which is based on this analysis. We then show how it can be applied to specific learning tasks and present results of nu- merical experiments showing that our method achieves state-of-the-art results for several learning tasks in quantum control with discrete and continouous control parameters.

Moritz August, Xiaotong Ni

We utilize machine learning models which are based on recurrent neural networks to optimize dynamical decoupling (DD) sequences. DD is a relatively simple technique for suppressing the errors in quantum memory for certain noise models. In numerical simulations, we show that with minimum use of prior knowledge and starting from random sequences, the models are able to improve over time and eventually output DD-sequences with performance better than that of the well known DD-families. Furthermore, our algorithm is easy to implement in experiments to find solutions tailored to the specific hardware, as it treats the figure of merit as a black box.