[Paper Review] Superlinear Amplitude Amplification

Lov K. Grover|ArXiv.org|Jun 2, 2008

Quantum Computing Algorithms and Architecture1 references3 citations

TL;DR

This paper introduces a novel quantum search algorithm that achieves superlinear (quadratic) amplitude amplification by redefining the core unitary operation to $ V = -U^{-1}I_t I_s U $, enabling the target state amplitude to grow quadratically with iterations. Unlike standard amplitude amplification, this method operates in higher-dimensional Hilbert spaces and offers faster initial amplification under specific conditions, though with a limited dynamic range.

ABSTRACT

Quantum search/amplitude amplification algorithms are designed to be able to amplify the amplitude in the target state linearly with the number of operations. Since the probability is the square of the amplitude, this results in the success probability rising quadratically with the number of operations. This paper presents a new kind of quantum search algorithm in which the amplitude of the target state, itself increases quadratically with the number of operations. However, the domain of applications of this is much more limited than standard amplitude amplification.

Motivation & Objective

• To develop a quantum algorithm that achieves superlinear (quadratic) growth in target state amplitude, surpassing the linear amplitude growth of standard amplitude amplification.
• To identify the conditions under which this quadratic amplification is possible, particularly in relation to the structure of the unitary transformation and initial state overlaps.
• To analyze the limitations of the method, especially its restricted dynamic range and higher-dimensional operational requirements compared to standard two-dimensional rotations.
• To explore the potential integration of this algorithm with existing quantum algorithms, such as the standard search algorithm or fixed-point amplitude amplification, to enhance robustness and amplification speed in specific regimes.

Proposed method

• The algorithm redefines the basic transformation as $ V = -U^{-1}I_t I_s U $, replacing the standard $ -UI_sU^{-1}I_tU $, to enable quadratic amplitude growth.
• The method relies on the recursive evolution of matrix elements, particularly $ V_{ts} = 2U_{ss}U_{ts}^* + 2U_{tt}^*U_{st} $, which depends on multiple matrix elements, not just $ U_{ts} $, indicating higher-dimensional dynamics.
• It assumes $ U_{ss} o 1 $ and $ U_{tt} o 1 $, allowing $ U_{ts} $ to grow by a factor of $ 4^i $ over $ i $ iterations, leading to quadratic scaling.
• The analysis uses a continuous limit approximation of the discrete recursion, leading to a differential equation system that models amplitude evolution over time.
• The dynamics are analyzed iteratively using three parameters: amplitude in the target state, source state, and other states, with average amplitude $ A(t) $ as a key variable.
• The system is solved in the continuous limit, yielding the target amplitude as $ rac{1}{2} - rac{1}{2} anig(rac{2 ilde{2}t}{ ilde{N}}ig) $, showing quadratic initial growth.

Experimental results

Research questions

• RQ1 Can amplitude amplification be achieved with quadratic, rather than linear, growth in the target state amplitude?
• RQ2 What unitary transformation structure enables this quadratic amplification, and how does it differ from standard amplitude amplification?
• RQ3 Why is this method restricted to a limited dynamic range, and what conditions must be met for it to function effectively?
• RQ4 How does the algorithm’s behavior differ from standard quantum search in terms of Hilbert space dimensionality and state vector evolution?
• RQ5 Can this method be combined with existing quantum algorithms to improve performance in specific application domains?

Key findings

• The target state amplitude grows quadratically with the number of iterations in the initial phase, specifically as $ rac{2t^2}{N} $, when $ t $ is small.
• The number of iterations required to reach unit amplitude is $ rac{ au ilde{N}}{2 ilde{2}} $, which is $ ilde{2} $ times more than the standard search algorithm’s $ rac{ au ilde{N}}{2} $, indicating a trade-off in speed versus robustness.
• The algorithm requires more than two dimensions to operate, as the recursion condition $ V_{ts} = 2U_{ss}U_{ts}^* + 2U_{tt}^*U_{st} $ cannot be satisfied in a two-dimensional Hilbert space due to unitarity constraints.
• When $ U $ is the inversion about average operation, $ U_{ts} = rac{2}{N} $, and the amplitude grows quadratically, but the method is only effective while $ U_{ss} o 1 $, which holds for approximately $ rac{ ilde{1}}{ ilde{4}} ilde{1} $ iterations.
• The method is not a two-dimensional rotation, unlike standard amplitude amplification, and thus cannot be described by a simple rotation in a plane, confirming its higher-dimensional nature.
• The algorithm may be useful as a pre-processing step in hybrid quantum-classical algorithms, such as in amplitude amplification pipelines, to boost initial amplitudes before applying more robust but slower methods.