%PDF-1.5 % 1 0 obj << /S /GoTo /D (section.1) >> endobj 4 0 obj (Introduction) endobj 5 0 obj << /S /GoTo /D (section.2) >> endobj 8 0 obj (Background) endobj 9 0 obj << /S /GoTo /D (subsection.2.1) >> endobj 12 0 obj (Model setting) endobj 13 0 obj << /S /GoTo /D (subsection.2.2) >> endobj 16 0 obj (False discovery rate control and the knockoff filter) endobj 17 0 obj << /S /GoTo /D (section.3) >> endobj 20 0 obj (Knockoffs inference for deep neural networks) endobj 21 0 obj << /S /GoTo /D (section.4) >> endobj 24 0 obj (Simulation studies) endobj 25 0 obj << /S /GoTo /D (subsection.4.1) >> endobj 28 0 obj (Model setups and simulation settings) endobj 29 0 obj << /S /GoTo /D (subsection.4.2) >> endobj 32 0 obj (Simulation results) endobj 33 0 obj << /S /GoTo /D (section.5) >> endobj 36 0 obj (Real data analysis) endobj 37 0 obj << /S /GoTo /D (subsection.5.1) >> endobj 40 0 obj (Application to HIV-1 data) endobj 41 0 obj << /S /GoTo /D (subsection.5.2) >> endobj 44 0 obj (Application to gut microbiome data) endobj 45 0 obj << /S /GoTo /D (section.6) >> endobj 48 0 obj (Conclusion) endobj 49 0 obj << /S /GoTo /D [50 0 R /Fit] >> endobj 55 0 obj << /Length 2287 /Filter /FlateDecode >> stream xXK6ϯ` |9v쬝ddf+sHHBL43m>کڽl@$|{C + X2*`IPn`ADr 7?ݼy&Β8GYfIܗo7J5~f&BՌSU |TkSv&~$ϲ`r ߬Ua44~q4)xdw]@I>FiPwa%$IRHBYЊ_7 쏄'$8D8eΑFb.2Os>ȍl{\ɅЪ.TS;NZox>lXy=/ACu}q07q4wb_g#ior]¥zs5iHT颵ȈHXQeY@Jd