RESUMO
A major archetype of artificial intelligence is developing algorithms facilitating temporal efficiency and accuracy while boosting the generalization performance. Even with the latest developments in machine learning, a key limitation has been the inefficient feature extraction from the initial data, which is essential in performance optimization. Here, we introduce a feature extraction method inspired by energy-entropy relations of sensory cortical networks in the brain. Dubbed the brain-inspired cortex, the algorithm provides convergence to orthogonal features from streaming signals with superior computational efficiency while processing data in a compressed form. We demonstrate the performance of the new algorithm using artificially created complex data by comparing it with the commonly used traditional clustering algorithms, such as Birch, GMM, and K-means. While the data processing time is significantly reduced-seconds versus hours-encoding distortions remain essentially the same in the new algorithm, providing a basis for better generalization. Although we show herein the superior performance of the cortical coding model in clustering and vector quantization, it also provides potent implementation opportunities for machine learning fundamental components, such as reasoning, anomaly detection and classification in large scope applications, e.g., finance, cybersecurity, and healthcare.
RESUMO
The extreme temperatures and uneven distribution of rainfall associated with climate change are expected to affect agricultural productivity and food security. A study was conducted to evaluate the impact of climate change on wheat in southeastern regions of Turkey. The CERES-wheat crop simulation model was calibrated and evaluated with data from eight surveyed farms. The four farms were used for calibration and four for evaluation. Climate change scenarios were developed for the middle (2036-2065) and late 21st century (2066-2095) under representative concentration pathways (RCPs 4.5 and 8.5) for study sites in Islahiye and Nurdagi. Model calibration results showed a good agreement between observed and simulated yield with only a 1 to 11% range of error. The model evaluation results showed good fit between observed and simulated values of all parameters with % error ranged from 0.51 to 13.3%. Future climate change projections showed that maximum temperature (Tmax) will increase between 1.6 °C (RCP4.5) and 2.3 °C (RCP8.5), while minimum temperature (Tmin) will increase between 1.0 °C (RCP4.5) and 1.5 °C (RCP8.5) for mid-century. At the end of the century, Tmax is projected to increase from 2 °C (RCP4.5) to 4 °C (RCP8.5) and Tmin from 1.3 °C (RCP4.5) to 3.1 °C (RCP8.5). Climate change impacts results showed that future rise in temperature will reduce wheat yield by 16.3% in mid-century and 16.8% at the end of the century at Islahiye and for Nurdagi, while 13.0% in mid and 14.4% end of the century. The use of climate and crop modeling technique provides useful information in evaluating the climate change impacts and may assist stakeholders to make decisions to overcome the negative impacts in the near and long term.
