Title: Universal quantum algorithmic cooling on a quantum computer

Abstract: Quantum cooling, a deterministic process that drives any state to the lowest eigenstate, has been widely used from studying ground state properties of chemistry and condensed matter quantum physics, to general optimization problems. However, the cooling procedure is generally non-unitary, hence its realization on a quantum computer either requires deep circuits or assumes specific input states with variational circuits. Here, we propose universal quantum cooling algorithms that overcome these limitations. By utilizing a dual phase representation of decaying functions, we show how to universally and deterministically realize a general cooling procedure with shallow quantum circuits. We demonstrate its applications in cooling an arbitrary input state with known ground state energy, corresponding to satisfactory, linear algebra tasks, and quantum state compiling tasks, and preparing unknown eigenvalues and eigenstates, corresponding to quantum many-body problems. Compared to quantum phase estimation, our method uses only one ancillary qubit and much shallower circuits, showing exponential improvement of the circuit complexity with respect to the final state infidelity. We numerically benchmark the algorithms for the $8$-qubit Heisenberg model and verify its feasibility for accurately finding eigenenergies and obtaining eigenstate measurements. Our work paves the way for efficient and universal quantum algorithmic cooling with near-term as well as universal fault-tolerant quantum devices.