{ "id": "2311.18313", "version": "v1", "published": "2023-11-30T07:31:36.000Z", "updated": "2023-11-30T07:31:36.000Z", "title": "Automatic Implementation of Neural Networks through Reaction Networks -- Part I: Circuit Design and Convergence Analysis", "authors": [ "Yuzhen Fan", "Xiaoyu Zhang", "Chuanhou Gao", "Denis Dochain" ], "categories": [ "math.DS", "cs.LG", "cs.NE" ], "abstract": "Information processing relying on biochemical interactions in the cellular environment is essential for biological organisms. The implementation of molecular computational systems holds significant interest and potential in the fields of synthetic biology and molecular computation. This two-part article aims to introduce a programmable biochemical reaction network (BCRN) system endowed with mass action kinetics that realizes the fully connected neural network (FCNN) and has the potential to act automatically in vivo. In part I, the feedforward propagation computation, the backpropagation component, and all bridging processes of FCNN are ingeniously designed as specific BCRN modules based on their dynamics. This approach addresses a design gap in the biochemical assignment module and judgment termination module and provides a novel precise and robust realization of bi-molecular reactions for the learning process. Through equilibrium approaching, we demonstrate that the designed BCRN system achieves FCNN functionality with exponential convergence to target computational results, thereby enhancing the theoretical support for such work. Finally, the performance of this construction is further evaluated on two typical logic classification problems.", "revisions": [ { "version": "v1", "updated": "2023-11-30T07:31:36.000Z" } ], "analyses": { "keywords": [ "neural network", "reaction network", "bcrn system achieves fcnn", "circuit design", "system achieves fcnn functionality" ], "note": { "typesetting": "TeX", "pages": 0, "language": "en", "license": "arXiv", "status": "editable" } } }