This paper introduces a novel causal attribution framework for supply chain simulations that combines Shapley values with Gaussian process emulators to decompose simulation outputs into individual input effects. The approach addresses the challenge of explaining complex simulation outputs by quantifying the contribution of each input feature. Experiments on synthetic and real-world supply chain data demonstrate the framework's ability to efficiently identify root causes of anomalies.

This paper investigates hardware-software co-design techniques to improve energy efficiency in large-scale deep learning training on NVIDIA, AMD, and emerging GPU architectures. It focuses on memory-level and kernel-level optimizations, including specialized tensor cores, memory optimization methods, mixed-precision arithmetic, and energy-aware scheduling. The study demonstrates that co-design can significantly improve training efficiency and reduce the carbon footprint of AI, supported by case studies from companies like Meta, Google, and Amazon.