const int k; // both users maximise the expected reward before their battery becomes empty <>max=? (R{"r1"}[F e1=0] + R{"r2"}[F e2=0]) // both users maximise the expected reward for the first k steps <>max=? (R{"r1"}[C<=k] + R{"r2"}[C<=k])