We propose an incomplete algorithm for Maximum Satisfiability (MaxSAT) specifically designed to run on neural network accelerators such as GPUs and TPUs. Given a MaxSAT problem instance in conjunctive normal form, our procedure constructs a Restricted Boltzmann Machine (RBM) with an equilibrium distribution wherein the probability of a Boolean assignment is exponential in the number of clauses it satisfies. Block Gibbs sampling is used to stochastically search the space of assignments with parallel Markov chains. Since matrix multiplication is the main computational primitive for block Gibbs sampling in an RBM, our approach leads to an elegantly simple algorithm (40 lines of JAX) well-suited for neural network accelerators. Theoretical results about RBMs guarantee that the required number of visible and hidden units of the RBM scale only linearly with the number of variables and constant-sized clauses in the MaxSAT instance, ensuring that the computational cost of a Gibbs step scales reasonably with the instance size. Search throughput can be increased by batching parallel chains within a single accelerator as well as by distributing them across multiple accelerators. As a further enhancement, a heuristic based on unit propagation running on CPU is periodically applied to the sampled assignments. Our approach, which we term RbmSAT, is a new design point in the algorithm-hardware co-design space for MaxSAT. We present timed results on a subset of problem instances from the annual MaxSAT Evaluation's Incomplete Unweighted Track for the years 2018 to 2021. When allotted the same running time and CPU compute budget (but no TPUs), RbmSAT outperforms other participating solvers on problems drawn from three out of the four years' competitions. Given the same running time on a TPU cluster for which RbmSAT is uniquely designed, it outperforms all solvers on problems drawn from all four years.
Northeast Normal University(东北师范大学);Institute of Software, Chinese Academy of Sciences(中国科学院软件研究所);Beijing University of Aeronautics and Astronautics(北京航空航天大学)
A general optimization solver based on OP-to-MaxSAT reduction
School of Software Engineering, South China University of Technology(华南理工大学软件学院);School of Software Engineering, Sun Yat-sen University(中山大学软件学院);Department of Mathematics, College of Science, Shantou University(汕头大学理学院数学系)
Learning Linear Temporal Properties from Noisy Data: A MaxSAT Approach
Ecole Polytechnique(巴黎综合理工学院);Max Planck Institute for Software Systems(马克斯·普朗克软件系统研究所);University of Texas at Austin(德克萨斯大学奥斯汀分校);Arizona State University(亚利桑那州立大学)
Solving the multiplication problem of a large language model system using a graph-based method
Firat University(菲拉特大学);Erzurum Technical University(埃尔祖鲁姆技术大学);University of Southern Queensland(南昆士兰大学);National Heart Centre Singapore(新加坡国家心脏中心);Duke-NUS Medical School(杜克-新加坡国立大学医学院);University of New England(新英格兰大学);University of Technology Sydney(悉尼科技大学)
Neural Approaches to SAT Solving: Design Choices and Interpretability
Czech Institute of Informatics, Robotics and Cybernetics, Czech Technical University in Prague(捷克布拉格捷克技术大学信息学、机器人学与控制论研究所);University of Ostrava(俄斯特拉发大学)
Learning Solving Procedure for Artificial Neural Network
Inha University(仁荷大学)
人们预计,通过集成表示学习和复杂推理的系统的出现,将实现向真正的人工智能的进展(LeCun et al. 2015)。针对这一预测,已经进行了在人工神经网络中实现冯·诺依曼计算机的符号推理的研究(Graves et al. 2016; Graves et al. 2014; Reed et al. 2015)。然而,这些研究在实现神经-符号积分方面有许多局限性(Jaeger. 2016年)。在这里,我们提出了一个新的学习范式:学习解决程序(LSP),学习解决复杂问题的程序。这不仅仅是通过学习输入-输出数据来实现的,而是通过通过求解过程的学习算法来实现的,该求解过程获得作为给定输入问题的任务序列的输出。LSP神经网络系统不仅学习简单的加法和乘法问题,而且学习复杂问题的算法,如复杂的算术表达式,排序和汉诺塔。为了实现这一点,LSP神经网络结构由递归组合的深度神经网络和长短期记忆组成。通过实验,我们证明了LSP的效率和可扩展性及其作为一种复杂推理机制的有效性。
Modern analog computing for solving differential and matrix equations
Institute for Artificial Intelligence, School of Integrated Circuits, Peking University, Beijing Advanced Innovation Center for Integrated Circuits(人工智能研究院,集成电路学院,北京大学,北京集成电路先进创新中心);Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano(电子、信息与生物工程系,米兰理工大学);IBM Research Europe(IBM欧洲研究院)