ChemReservoir -- An Open-Source Framework for Chemically-Inspired Reservoir Computing
Reservoir computing is a type of a recurrent neural network, mapping the inputs into higher dimensional space using fixed and nonlinear dynamical systems, called reservoirs. In the literature, there are various types of reservoirs ranging from in-silico to in-vitro. In cheminformatics, previous studies contributed to the field by developing simulation-based chemically inspired in-silico reservoir models. Yahiro used a DNA-based chemical reaction network as its reservoir and Nguyen developed a DNA chemistry-inspired tool based on Gillespie algorithm. However, these software tools were designed mainly with the focus on DNA chemistry and their maintenance status has limited their current usability. Due to these limitations, there was a need for a proper open-source tool. This study introduces ChemReservoir, an open-source framework for chemically-inspired reservoir computing. In contrast to the former studies focused on DNA-chemistry, ChemReservoir is a general framework for the construction and analysis of chemically-inspired reservoirs, which also addresses the limitations in these previous studies by ensuring enhanced testing, evaluation, and reproducibility. The tool was evaluated using various cycle-based reservoir topologies and demonstrated stable performance across a range of configurations in memory capacity tasks.
Code (1)
Methods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Aligning Molecular Graph Explanations with Chemical Identity via InChIfied Invariants
Obtaining consistent explanations for machine learning on molecular graphs requires predictions and attributions to be aligned with chemical identity. However, chemically equivalent drawings of the same molecule can indu…
SpinCastML an Open Decision-Making Application for Inverse Design of Electrospinning Manufacturing: A Machine Learning, Optimal Sampling and Inverse Monte Carlo Approach
Electrospinning is a powerful technique for producing micro to nanoscale fibers with application specific architectures. Small variations in solution or operating conditions can shift the jet regime, generating non Gauss…
FragmentNet: Adaptive Graph Fragmentation for Graph-to-Sequence Molecular Representation Learning
Molecular property prediction uses molecular structure to infer chemical properties. Chemically interpretable representations that capture meaningful intramolecular interactions enhance the usability and effectiveness of…
Graph-to-SequenceMolecular Property Predictionmolecular representationProperty Prediction+2GraphAF: a Flow-based Autoregressive Model for Molecular Graph Generation
Molecular graph generation is a fundamental problem for drug discovery and has been attracting growing attention. The problem is challenging since it requires not only generating chemically valid molecular structures but…
Density EstimationDrug DiscoveryGraph GenerationMolecular Graph Generation+2GEOM-Drugs Revisited: Toward More Chemically Accurate Benchmarks for 3D Molecule Generation
Deep generative models have shown significant promise in generating valid 3D molecular structures, with the GEOM-Drugs dataset serving as a key benchmark. However, current evaluation protocols suffer from critical flaws,…
3D Molecule GenerationBenchmarkingvalid