<p>The use of optimization algorithms is essential to train neural networks effectively. The usage of a combination of two different optimizers is proposed in this method that, used together, can perform single optimizers in various problems. The objective of this study is to develop a comprehensive framework that combines meta-optimization techniques with traditional optimization methods to enhance performance metrics such as convergence speed, solution quality, and robustness against local minima. The researcher presents ATMO (Adaptive Meta Optimizers), recent deep neural network (DNN) models that are combined with a new optimizer that combines two different optimizers simultaneously. Meta-optimizer considers all of their contributions at once, rather than trying to enhance each one separately. Using a wide range of techniques, DNNs have been the subject of numerous investigations on text and image categorization. Results from these tests demonstrated that the suggested Adaptive Meta Optimizer outperforms individual optimizers. These experiments show that adequate training is achieved by considering a variety of factors such as Keras standard learning rate and convergence, and degradation on the CIFAR-100, CIFAR-10, and MNIST datasets, and the results are obtained using Python.</p>