E. Küçüksille

Time Series Forecasting of Greenhouse Gas Emissions Using Deep Learning

Greenhouse gas (GHG) emissions remain a primary driver of global climate change, and accurate forecasting is critical for evaluating climate policies and supporting sustainable development goals. This study conducts a time series analysis in Python using the EDGAR dataset for 1970–2023, explicitly adopting a sectoral scope that covers Agriculture, Buildings, Fuel Exploitation, Industrial Combustion, Power Industry, Processes, Transport, and Waste. Long Short-Term Memory (LSTM) models were developed and evaluated via rolling/expanding-window backtesting, while Monte Carlo Dropout (MCD) was applied to quantify predictive uncertainty. Out-of-distribution (OOD) tests were further used to assess generalization under distributional shifts, and early stopping with learning-rate scheduling was employed to mitigate overfitting. While the LSTM captures the dominant long-term upward trend, its out-of-sample performance is constrained under regime changes and shocks (Test RMSE = 1228.66; MAE = 991.35; R2=−1.0020) and it underperforms a naïve benchmark. Nested rolling-origin results also indicate rapidly increasing errors at longer horizons. In contrast, a Transformer specification improves performance over the last five years (RMSE = 777.21; MAE = 663.39; R2=0.1989). Sectoral findings suggest that short-term declines around 2020 are concentrated in transport and energy-related sectors, while agriculture and industrial/process-related emissions remain relatively stable.

From DevOps to XOps: an agent-driven reference architecture for autonomous enterprise operations

Enterprise adoption of machine learning has fragmented operations into specialised disciplines—DataOps, MLOps, AIOps—creating silos that impede unified governance. We propose XOps, a five-layer reference architecture integrating PlatformOps, DataOps, MLOps and AIOps beneath an Agentic Orchestration layer with Policy-as-Code governance, together with a continuous-time Markov chain model quantifying the availability effect of agent-driven remediation. Two case studies evaluate the architecture under controlled, synthetic conditions. For a self-healing payment gateway, 250 live executions of the reasoning graph against a hosted language model and a live policy engine yield 85.6% plan-level action consistency and 99.6% fault classification accuracy, with 12.4% of plans rejected by the policy gate and escalated to a human and no policy-violating action authorised for execution; the pipeline’s 3.3-minute recovery time is a simulated latency budget rather than a cluster measurement. For a predictive-maintenance application on NASA C-MAPSS data, autonomous drift detection and retraining sustain $$R^2 = 0.74$$ against 0.29 for an equivalent static model, measured on engines reserved entirely from retraining. An indicative cost analysis suggests an approximately 70% reduction in expected monthly operational cost. Within this scope the results support the feasibility of agent-driven operations rather than establishing production-scale performance.

SeqGPT-RL: a reinforcement-aligned hybrid intelligent system for Turkish technical-support dialogs