1: \begin{abstract}

2: This paper is concerned with convergence analysis \ill{for the mirror descent (MD) method, a well-known algorithm in convex optimization.}

3: %of the well-known mirror descent (MD) method.

4: An analysis framework via integral quadratic constraints (IQCs) is {constructed} %constructed

5:  to analyze the convergence rate of the MD method with strongly convex objective functions in both continuous-time and discrete-time.

6: %\todoiny{I removed the ``-'' because I think ``continuous-time'' is more like an adjective.}

7: We formulate the problem of finding convergence rates of the MD algorithms into feasibility problems of linear matrix inequalities (LMIs) in both schemes.

8: \icl{In particular, in continuous-time}, we show that \icl{the Bregman divergence function, which is commonly used as a Lyapunov function for this algorithm,} is a special case of \icl{the class of Lyapunov functions} \icl{associated with the} Popov criterion, \icl{when the latter is applied to an appropriate \ill{reformulation} of the problem}. Thus, applying the Popov criterion \icl{and its combination with other IQCs,}

9: %yields a less conservative

10: \ill{can lead to convergence rate bounds with reduced conservatism.} 

11: % ared to using only the Bregman function \icl{as a Lyapunov function}.

12: %\icl{We also show that in %the

13: %discrete-time, the} unconstrained MD method and its projected variant have the same upper bound of convergence rates under the IQC \icl{framework developed in the paper,} \ml{which saves the effort of analyzing extra projection dynamics for the constrained MD algorithm.}

14: %%which saves the effort in the analysis.

15: %\todoing{\st{"saves the effort in the analysis" not very clear needs rephrasing}}

16: \ml{We also \il{illustrate via examples that the %provide a tight

17: convergence rate bounds derived \ill{can be} tight.}}

18: %in the numerical example for the discrete-time MD algorithm.}

19: \todoing{\st{Last sentence not very clear}}

20: \end{abstract}

21: