Our standards

هذه الصفحة متاحة أيضًا باللغة العربية.

التبديل إلى العربية

Esta página también está disponible en español.

Cambiar a español

Cette page est également disponible en français.

Passer en français

Bu sayfa Türkçe olarak da mevcut.

Türkçe sürüme geç
HABER PAPER / NEWSROOM

Artificial intelligence — Page 5

Artificial intelligence research, applications, data infrastructure and verified corporate announcements. Page 5.

122 stories

Browse topics
1 min read

EnergyEvent / data period:

Best AI model for wind-turbine faults varied by fault type

The 13 October Scientific Reports preview compares SVM and a neural network with a Random Forest baseline for wind-turbine fault detection. Its evaluation concerns simulation conditions.

Source: Scientific Reports, Data-driven fault detection in wind turbine systems using support vector machines and artificial neural networks
1 min read

BusinessEvent / data period:

Ottawa team reports pathology AI that segments slides with fewer labels

SlideFiLM feeds whole-slide context into tissue-image tiles to segment kidney and vascular structures. Supplementary tables report results when using just 20% of labeled training data.

Source: npj Imaging, Data-efficient pathology whole-slide image segmentation with slide-level conditioning
1 min read

BusinessEvent / data period:

32 UAE experts develop AI projects for national priorities

Thirty-two Emirati specialists in the National Experts Programme’s artificial intelligence track are working on projects addressing national priorities across sectors. Final presentations are scheduled for January 2027. Panels of 20 leaders, programme alumni and investors will assess the projects.

Source: Emirates News Agency (WAM)
1 min read

EnergyEvent / data period:

AI model tests sensor placement for power-flow estimation

Researchers affiliated with Laval University evaluated a neural network that estimates power flows with fewer phasor measurement units on the IEEE 30-bus test grid. Results show that where sensors are placed can matter more than using the same number of devices.

Source: arXiv, Learning-Based Power Flow Estimation with Sparse PMU Deployment in Power Systems (arXiv:2610.13683)
1 min read

BusinessEvent / data period:

Thermal and visible cameras jointly detect wheel tracks on snowy roads

Michigan Tech researchers presented a self-supervised model that combines RGB and infrared images to identify existing wheel tracks on snow-covered roads. Tests on the ThermalTrack dataset found fewer missed tracks than with RGB alone.

Source: arXiv, Tracking the Invisible: Self-Supervised Wheel Track Detection with Thermal-RGB Fusion (arXiv:2610.13872)
1 min read

BusinessEvent / data period:

Skipping some Transformer blocks cut forecasting latency in tests

A Polytechnique Montréal study tested skipping Transformer blocks during inference in Chronos-Bolt and TimesFM 2.5. Moderate skipping largely preserved forecasting quality across four benchmark datasets while lowering A100 latency.

Source: arXiv, How Much Computation Does Every Forecast Need? Structured Block Skipping in Two Time-Series Foundation Models (arXiv:2610.13208)
1 min readWith data

BusinessEvent / data period:

AMD introduces Helion rack design for agentic AI

The open blueprint offers CPU and GPU configurations; partner-developed systems are planned for availability in November.

Source: AMD Investor Relations
1 min read

AgricultureEvent / data period:

Solar-powered LSTM irrigation system for Egypt tested in simulation

A Scientific Reports study built a MATLAB smart-irrigation simulation for Egypt’s New Delta using NASA POWER environmental data. It classified pump states with 96.11% accuracy; water and energy savings are model estimates, not field measurements.

Source: Scientific Reports, A MATLAB-based AI-IoT simulation framework for LSTM-driven smart solar irrigation and renewable energy management
1 min read

BusinessEvent / data period:

Language models can miss atomic geometry in materials property prediction

Research co-authored by Intel Labs reports that language models learn composition labels better than spatial coordinates. Geometry-focused graph networks performed better on these tasks.

Source: Nature Computational Science, Less can be more for predicting properties with large language models
HABER PAPER

Five editions. One wider view.

A multilingual perspective on the global economy.