Default acceptance Parameter probably wrong in dual annealing

[echo5india]

Hello,

I think the parameter 'accept' in the dual annealing function has a wrong default value and wrong acceptable bounds. As explained in this paper Tsallis C, Stariolo DA. Generalized Simulated Annealing. Physica A, 233, 395-406 (1996), the acceptance parameter should be above 1 (probably slightly above to be close to classic simulated annealing).

The descrition for this parameter is actually :

"""Parameter for acceptance distribution. It is used to control the probability of acceptance. The lower the acceptance parameter, the smaller the probability of acceptance. Default value is -5.0 with a range (-1e4, -5]."""

Without an acceptable parameter here the algorithms never accepts states of higher energy and the algorithms does not behave as expected.

many thanks

[leostimpfle]

The paper does not state that the acceptance parameter qA should be larger than 1. In fact, on p. 402 the authors state that

it is found that the convergence is faster for values of qA < I

and on p. 403 that

the best performing cases are those with visiting parameter qv ~- 2.7 and the acceptance parameter qA as negative as possible.

However, it would be useful to include documentation on how the range (-5, -1e4] of the acceptance parameter has been chosen.