Abstract
Although spatial locality is explicit in the Heisenberg picture of quantum dynamics, spatial locality is not explicit in the Schrödinger picture equations of motion. The gauge picture is a modification of Schrödinger's picture such that locality is explicit in the equations of motion. In order to achieve this explicit locality, the gauge picture utilizes (1) a distinct wavefunction associated with each patch of space, and (2) time-dependent unitary connections to relate the Hilbert spaces associated with nearby patches. In this work, we show that by adding an additional spatially-local term to the gauge picture equations of motion, we can effectively interpolate between the gauge and Schrödinger pictures, such that when this additional term has a large coefficient, all of the gauge picture wavefunctions approach the Schrödinger picture wavefunction (and the connections approach the identity).
Introduction
Standard quantum mechanics typically employs the Schrödinger picture, where a global wavefunction evolves in time, or the Heisenberg picture, where observables evolve. A significant challenge in the Schrödinger picture is that spatial locality is not manifest. While physical interactions are local, the wavefunction is a global object describing the entire system, often obscuring the local nature of dynamics in the mathematical formalism.
The recently proposed Gauge picture addresses this by assigning distinct wavefunctions to spatial "patches" and using unitary connections to relate them. However, this approach "scrambles" local wavefunctions, making them look different from the standard Schrödinger state. This research introduces a Modified Gauge Picture that bridges these two descriptions, allowing for explicitly local equations while forcing local wavefunctions to remain "unscrambled" and consistent with the global Schrödinger state.
Methods
The authors modify the Gauge picture equations of motion for local wavefunctions ($|\psi_I\rangle$) and connections ($U_{IJ}$) by adding a new term, $\gamma X_I$, to the generator of time evolution.
- The Interpolator ($\gamma$): A real-valued constant. At $\gamma = 0$, the system follows the standard Gauge picture. As $\gamma$ increases, it drives the connections $U_{IJ}$ toward the identity matrix.
- The Driving Term ($X_I$): A Hermitian operator designed to minimize connection deviation from the identity without altering physical predictions.
The framework was validated using numerical simulations of the 1D transverse-field Ising model (TFIM) for system sizes $L = 5$ to $10$. A modified 4th-order Runge-Kutta method was used to maintain unitary constraints during the simulation.
Key Insights and Results
The study reveals that the $\gamma$ term successfully acts as a bridge. For large $\gamma$, local Gauge picture wavefunctions converge to the global Schrödinger wavefunction. However, a critical finding is the exponential cost of locality: to maintain this convergence as system size $L$ grows, $\gamma$ must scale exponentially with $L$.
Scaling Laws and Numerical Stability
- Scaling Law: The deviation $S_{IJ}$ follows $S_{IJ}(\infty) \sim \gamma^{-2} e^{aL+b}$, confirming the exponential dependence on system size.
- Divergence of Squiggles: For small $\gamma$, the deviation exhibits "squiggles" or oscillations. The onset time $t_s$ follows $t_s^2 \propto 1/(\gamma - \gamma_0)$, allowing prediction of when asymptotic behavior breaks down.
- Numerical Chaos: The non-linear nature of Gauge picture equations leads to numerical chaos, where small integration errors grow exponentially. This makes long-time simulations more difficult than in the linear Schrödinger picture.
Conclusion
The requirement for $\gamma$ to scale exponentially with $L$ limits the practical application of this interpolation for very large systems. Additionally, the non-linear instability of the Gauge picture presents a significant numerical hurdle. Future work may focus on stabilizing these simulations and extending the modified Gauge picture to higher-dimensional lattices, where the scaling laws may reveal further complexities of local-global equivalence in quantum dynamics.
Summary updated