Thermal bootstrap bounds improve large-N matrix model energies

What the study found: The study reports improved thermal bootstrap bounds for matrix quantum mechanics, specifically for the thermal energies of large-N one-matrix and two-matrix anharmonic oscillators. In the one-matrix case, the authors say the stricter bounds give an estimate of the first long string excited energy within 0.001% of the physical value and provide the first estimate of the first long string coupling coefficient from symmetry and self-consistency equations alone.

Why the authors say this matters: The authors present these results as an improvement to thermal bootstrapping methods for matrix quantum mechanics. They also note that, in the low-temperature limit, the one-matrix model can be interpreted through an effective theory of "long strings".

What the researchers tested: The paper applies thermal bootstrapping to large-N matrix models, including the one-matrix anharmonic oscillator and the two-matrix anharmonic oscillator. The thermal energies were bounded using the Quantum Information Conic Solver, and the authors report doing this without logarithmic relaxation.

What worked and what didn't: The bounding procedure worked for both the large-N one-matrix and two-matrix anharmonic oscillator models, according to the abstract. For the one-matrix model, the bounds were tight enough to estimate the first long string excited energy to within 0.001% of the physical value and to infer the first long string coupling coefficient; no failures or negative results are described in the abstract.

What to keep in mind: The abstract does not describe detailed limitations, experimental constraints, or failure cases. The claims are limited to the models and methods named in the abstract, so no broader applicability is stated there.

Key points

• The paper reports improved thermal bootstrap bounds for large-N matrix quantum mechanics.
• Thermal energies were bounded for one-matrix and two-matrix anharmonic oscillators using the Quantum Information Conic Solver.
• For the one-matrix model, the first long string excited energy was estimated within 0.001% of the physical value.
• The authors say they obtained the first estimate of the first long string coupling coefficient from symmetry and self-consistency equations alone.
• The abstract does not describe limitations or negative results.