1: \begin{abstract}

2: 		We present a fast update rule for variational block-\gls{sbl} methods.

3: 		Based on a variational Bayesian approximation, we show that iterative updates of \glspl{pdf} of the prior precisions and weights can be expressed as a nonlinear first-order recurrence from one estimate of the parameters of the proxy \glspl{pdf} to the next. In particular, for commonly used prior \glspl{pdf} such as Jeffrey's prior, the recurrence relation turns out to be a strictly increasing rational function.

4: 		This property is the basis for two important analytical results.

5: 		First, the determination of fixed points by solving for the roots of a polynomial.

6: 		Second, the determination of the limit of the prior precision after an infinite sequence of update steps. These results are combined into a simplified single-step check for convergence/divergence of each prior precision.

7: 		Consequently, our proposed criterion significantly reduces the computational complexity of the variational block-\gls{sbl} algorithm, leading to a remarkable two orders of magnitude improvement in convergence speed shown by simulations.

8: 		Moreover, the criterion provides valuable insights into the sparsity of the estimators obtained by different prior choices.

9: 	\end{abstract}

10: