The paper presents an all-optical spiking neurosynaptic network capable of self-learning. The authors demonstrate a photonic neuron device connected to photonic synapses via an integrated photonics network, forming a small-scale all-optical neurosynaptic system. This system can perform supervised and unsupervised learning tasks, such as pattern recognition. The network is designed using wavelength division multiplexing (WDM) techniques to achieve a scalable circuit architecture for photonic neural networks. The photonic system consists of 140 elements and successfully recognizes patterns directly in the optical domain. The use of phase-change materials (PCMs) in the synapses allows for hardware implementation of the integrate-and-fire functionality of neurons and the plastic weighting operation of synapses. The all-optical spiking neural network offers high speed and bandwidth, making it suitable for processing telecommunication and visual data. The paper also discusses the fabrication process and experimental setup for the photonic circuits, as well as the measurement setup for operating the all-optical neurons.The paper presents an all-optical spiking neurosynaptic network capable of self-learning. The authors demonstrate a photonic neuron device connected to photonic synapses via an integrated photonics network, forming a small-scale all-optical neurosynaptic system. This system can perform supervised and unsupervised learning tasks, such as pattern recognition. The network is designed using wavelength division multiplexing (WDM) techniques to achieve a scalable circuit architecture for photonic neural networks. The photonic system consists of 140 elements and successfully recognizes patterns directly in the optical domain. The use of phase-change materials (PCMs) in the synapses allows for hardware implementation of the integrate-and-fire functionality of neurons and the plastic weighting operation of synapses. The all-optical spiking neural network offers high speed and bandwidth, making it suitable for processing telecommunication and visual data. The paper also discusses the fabrication process and experimental setup for the photonic circuits, as well as the measurement setup for operating the all-optical neurons.