1: \begin{abstract}

2: For the problem of nonparametric regression of smooth functions, 

3: we {reconsider} and analyze a constrained variational approach, 

4: which we call the MultIscale Nemirovski-Dantzig (MIND) estimator.  

5: This can be viewed as a multiscale extension of the Dantzig selector

6:  (\emph{Ann. Statist.}, 35(6): 2313--51, 2009) based on early ideas of

7: Nemirovski (\emph{J. Comput. System Sci.}, 23:1--11, 1986). MIND

8: minimizes a homogeneous Sobolev norm under the constraint that 

9: the multiresolution norm of the residual is bounded by a universal threshold.

10: The main contribution of this paper is the derivation of

11: convergence rates of MIND with respect to $L^q$-loss, $1 \le q \le

12: \infty$, both almost surely and in expectation. {To this end,  we introduce} the method of

13: approximate source conditions. {For a

14: one-dimensional signal,} {these can be translated into} 

15: approximation properties of $B$-splines.

16: A remarkable consequence is that MIND attains almost

17: minimax optimal rates simultaneously for a large range of Sobolev and

18: Besov classes, {which provides certain adaptation.} 

19: Complimentary to the asymptotic analysis, we examine

20: the finite sample performance of MIND by numerical simulations.

21: \end{abstract}

22: