Some honeybees learn tasks better than others, and gene expression patterns in their brains may be associated with this difference in ability

Researchers analyzed DNA samples from foraging honeybees, hive workers, and their honey to understand plant-microbe interactions in honeybee ecology. The study found that these complementary sources provide distinct insights into the complex relationships between plants, microbes, and honeybees.

A recent study reveals that ORC2 subunit regulates epigenetics and gene expression by compacting chromatin and attracting repressive histone marks at some sites, but activating gene expression at others. This regulation also prevents CTCF binding at certain sites, leading to changes in chromatin structure and gene expression.

Researchers at UC Berkeley developed an AI-powered training method called Human-in-the-Loop Sample Efficient Robotic Reinforcement Learning (HiL-SERL) that enables robots to perform complicated tasks with precision and speed. With human feedback, robots learn from demonstrations and real-world attempts, achieving a 100% success rate in...

The brain uses reusable patterns of activity across tasks to build new behaviors, snapping together 'cognitive Legos' that can be flexibly combined. This compositionality enables the brain to quickly learn new information and switch between tasks, a feature current AI models struggle with.

Researchers have developed a framework called CHEEM that allows AI models to learn new tasks without losing performance on existing tasks. The framework improves adaptive intelligence by tailoring computational structure depending on task complexity.

Scientists at the University of Amsterdam have developed metamaterials that learn and adapt without a central brain, allowing them to change shape and perform advanced tasks. These 'smart' materials can forget old shapes and learn new ones, enabling them to evolve and perform complex tasks.