Breaking the Deadly Triad with a Target Network

                Shangtong Zhang 1 Hengshuai Yao 2 3 Shimon Whiteson 1

             Abstract                 ping methods construct update targets for an estimate by
                                  using the estimate itself recursively, which usually has lower
    The deadly triad refers to the instability of a re-     variance than Monte Carlo methods (Sutton, 1988). How-
    inforcement learning algorithm when it employs       ever, when an al ...