1: \begin{abstract}

2: We introduce a general form of sequential Monte Carlo algorithm defined

3: in terms of a parameterized resampling mechanism. We find that a suitably

4: generalized notion of the Effective Sample Size (ESS), widely used

5: to monitor algorithm degeneracy, appears naturally in a study of its

6: convergence properties. We are then able to phrase sufficient conditions

7: for time-uniform convergence in terms of algorithmic control of the

8: ESS, in turn achievable by adaptively modulating the interaction between

9: particles. This leads us to suggest novel algorithms which are, in

10: senses to be made precise, provably stable and yet designed to avoid

11: the degree of interaction which hinders parallelization of standard

12: algorithms. As a byproduct we prove time-uniform convergence of the

13: popular adaptive resampling particle filter.

14: 

15: \medskip{}

16: 

17: 

18: \emph{Keywords:} convergence; hidden Markov model; particle filters

19: \end{abstract}