Post-Selection Inference with HSIC-Lasso

       Tobias Freidling 1 Benjamin Poignard 2 3 Hector Climente-Gonzalez 3 Makoto Yamada 3 4

             Abstract                 The most straightforward remedy is sample splitting,
                                  which uses one part of the data for model selection and the
    Detecting influential features in non-linear and/or     other part for inference, cf. (Cox, 1975). However, this
    high-dimensional data is a challenging and in-        ap ...