I would consider a successful application of AlphaZero algorithm to imperfect information games to be a separate achievement, not a done deal. David Silver him

I would consider a successful application of AlphaZero algorithm to imperfect information games to be a separate achievement, not a done deal. David Silver himself, who led AlphaGo project, wrote in 2016 in "Deep Reinforcement Learning from Self-Play in Imperfect-Information Games" https://arxiv.org/abs/1603.01121 https://arxiv.org/abs/1603.01121 "While many machine learning methods have achieved near-optimal solutions to classical, perfect-information games, these methods fail to converge in imperfect-information games".