Evaluating Adaptive Neural Architectures for Real-Time Decision-Making in Autonomous Systems
Keywords:
adaptive neural networks, real-time inference, autonomous systems, dynamic computation, decision-makingAbstract
Real-time decision-making is critical for autonomous systems such as self-driving vehicles and robotic agents. This paper evaluates adaptive neural architectures designed to respond to changing environmental dynamics with high-speed processing and accurate inference. We propose a hybrid approach leveraging dynamic computational graphs and attention-driven modular networks. Empirical benchmarks demonstrate superior performance of adaptive architectures over fixed models across latency, accuracy, and generalization metrics. The study suggests adaptive networks significantly enhance autonomous systems’ capacity for timely and context-sensitive responses.
References
Figurnov, M., Collins, M. D., Zhu, Y., Zhang, L., Huang, J., Vetrov, D., & Salakhutdinov, R. (2017). Spatially adaptive computation time for residual networks. CVPR, 1039–1048.
Graves, A. (2016). Adaptive computation time for recurrent neural networks. arXiv preprint arXiv:1603.08983.
Han, S., Pool, J., Tran, J., & Dally, W. (2015). Learning both weights and connections for efficient neural networks. NIPS, 28.
Sheetal, J. (2022). Redefining resilience through architectural innovation and operational excellence in SAP HANA backup implementation on Microsoft Azure for scalable, secure and intelligent data protection. IACSE – International Journal of Scientific Computing, 3(1), 9–31. https://doi.org/10.5281/zenodo.17785816
Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., ... & Adam, H. (2017). MobileNets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861.
Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017). Densely connected convolutional networks. CVPR, 4700–4708.
Iandola, F. N., Han, S., Moskewicz, M. W., Ashraf, K., Dally, W. J., & Keutzer, K. (2016). SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size. arXiv preprint arXiv:1602.07360.
LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444.
Liu, H., Simonyan, K., & Yang, Y. (2018). DARTS: Differentiable architecture search. arXiv preprint arXiv:1806.09055.
Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A. A., Veness, J., Bellemare, M. G., ... & Hassabis, D. (2015). Human-level control through deep reinforcement learning. Nature, 518(7540), 529–533.
Pasumarthi, A. (2021). Intelligent system optimization and automation for large-scale SAP landscapes: A framework for predictive stability and operational efficiency. IACSE – Global Journal of Information Technology, 2(1), 8–19. https://doi.org/10.5281/zenodo.17826318
Pham, H., Guan, M. Y., Zoph, B., Le, Q. V., & Dean, J. (2018). Efficient neural architecture search via parameter sharing. ICML, 4095–4104.
Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., Van Den Driessche, G., ... & Hassabis, D. (2016). Mastering the game of Go with deep neural networks and tree search. Nature, 529(7587), 484–489.
Zoph, B., & Le, Q. V. (2017). Neural architecture search with reinforcement learning. ICLR.
