1: \begin{abstract}

2: 	We consider the non-parametric Poisson regression problem where the integer valued response $Y$ is the realization of a Poisson random variable with parameter $\lambda(X)$.

3: 	The aim is to estimate the functional parameter $\lambda$ from independent or weakly dependent observations $(X_1,Y_1),\ldots,(X_n,Y_n)$ in a random design framework.

4: 	

5: 	First we determine upper risk bounds for projection estimators on finite dimensional subspaces under mild conditions.

6: 	In the case of Sobolev ellipsoids the obtained rates of convergence turn out to be optimal.

7: 

8: 	The main part of the paper is devoted to the construction of adaptive projection estimators of $\lambda$ via model selection.

9: 	We proceed in two steps: first, we assume that an upper bound for $\Vert \lambda \Vert_\infty$ is known. 

10: 	Under this assumption, we construct an adaptive estimator whose dimension parameter is defined as the minimizer of a penalized contrast criterion.

11: 	Second, we replace the known upper bound on $\Vert \lambda \Vert_\infty$ by an appropriate plug-in estimator of $\Vert \lambda \Vert_\infty$.

12: 	The resulting adaptive estimator is shown to attain the minimax optimal rate up to an additional logarithmic factor both in the independent and the weakly dependent setup.

13: 	Appropriate concentration inequalities for Poisson point processes turn out to be an important ingredient of the proofs.

14: 	

15: 	We illustrate our theoretical findings by a short simulation study and conclude by indicating directions of future research.

16: \end{abstract}