Reverse Polish notation

Abstract: This paper develops an algorithm that extracts explanatory rules from microarray data, which we treat as time series, using genetic programming (GP) and fuzzy logic. Reverse polish notation is used (RPN) to describe the rules and to facilitate the GP approach. The algorithm also allows for the insertion of prior knowledge, making it possible to find sets of rules that include the relationships between genes already known. The algorithm proposed is applied to problems arising in the construction of gene regulatory networks, using two different sets of real data from biological experiments on the Arabidopsis thaliana cold response and the rat central nervous system, respectively. The results show that the proposed technique can fit data to a pre-defined precision even in situations where the data set has thousands of features but only a limited number of points in time are available, a situation in which traditional statistical alternatives encounter difficulties, due to the scarcity of time points.

Abstract: We present a novel Dynamic Differentiable Reasoning (DDR) framework for jointly learning branching programs and the functions composing them; this resolves a significant nondifferentiability inhibiting recent dynamic architectures. We apply our framework to two settings in two highly compact and data efficient architectures: DDRprog for CLEVR Visual Question Answering and DDRstack for reverse Polish notation expression evaluation. DDRprog uses a recurrent controller to jointly predict and execute modular neural programs that directly correspond to the underlying question logic; it explicitly forks subprocesses to handle logical branching. By effectively leveraging additional structural supervision, we achieve a large improvement over previous approaches in subtask consistency and a small improvement in overall accuracy. We further demonstrate the benefits of structural supervision in the RPN setting: the inclusion of a stack assumption in DDRstack allows our approach to generalize to long expressions where an LSTM fails the task.

Abstract: A computer-aided modelling tool, MoT, that assists the model developer in terms of model import, model translation, model analysis, model solution and model transfer without the user having to write any programming codes is presented. The main features of MoT are presented within the context of the work process related to various modelling activities during the life of a process. External models written in text-format and/or XML-format can be imported to MoT, which then translates and expands the model according to a Reverse Polish Notation algorithm. The translated model can be solved, after satisfying mathematical consistency requirements, equation by equation in the debug-mode or simultaneously in the solution-mode. The solvable model can also be exported through a model transfer feature to other simulation engines and/or external software. The use of MoT is highlighted through a number of interesting and illustrative modelling examples.

Abstract: We accelerate state space exploration for explicit-state model checking by executing complex operations on the graphics processing unit (GPU). In contrast to existing approaches enhancing model checking through performing parallel matrix operations on the GPU, we parallelize the breadth-first layered construction of the state space graph. For efficient processing, the input model is translated to the reverse Polish notation, resulting in a representation as an integer vector. The proposed GPU exploration algorithm then divides into two parallel stages. In the first stage, each state is replaced with a Boolean vector to denote which transitions are enabled. In the second stage, pairs consisting of replicated states and enabled transition IDs are copied to the GPU then all transitions are applied in parallel to produce the successors. Bitstate hashing is used as a Bloom filter to remove duplicates from the set of successors in RAM. The experiments show speed-ups of about one order of magnitude. Compared to state-of-the-art in multi-core model checking software, still advances remain visible.