All articles tagged with #neural networks
A microfluidics study shows E. coli can learn from past nutrient fluctuations by storing memory in ribosome subpopulations, influencing future growth. This history-dependent computation behaves like a gated recurrent neural network, offering a biological blueprint for continuous, energy-efficient learning with potential implications for medicine and AI design.
A startup called Mostik, with input from Russian mathematicians, is testing a method for AI models to communicate with one another without human language—a 'machine telepathy' approach that could speed collaboration between models.
A study of the feather moss Brachythecium rutabulum records long-term electrical activity in dense mats, uncovering fast, intermediate, and slow spike regimes and evidence of directional signal propagation. While not a human brain, the moss shows a rich electrophysiological repertoire that could underpin information processing and inspire biohybrid sensing applications, challenging the view of moss as a simple organism.
Tech writer explains Adrian de Wynter’s experiment building a tiny neural network inside Age of Empires II using goats to mimic a NAND gate and a 1-bit perceptron, showing that the same underlying LLM tech can appear human-like or not depending on the interface. The piece argues that anthropomorphism of AI is driven as much by presentation and marketing as by function, cites Ted Chiang’s Word analogy, and calls for clearer disclosures and alignment to avoid conflating a tool with consciousness.
In a math-grounded argument, the Slate piece contends that modern AI simply performs massive calculations and creates the illusion of consciousness, just as Muybridge’s rapid-fire photos suggested movement without life. By viewing AI as a sequence of mathematical operations, the author argues there is no mechanism by which current neural networks can achieve true self-awareness or continuous, sentient experience, regardless of how impressive their outputs may seem.
A Nature study shows subliminal learning: when a teacher model with a trait is used to generate data for distillation, a student can acquire that trait even if the data contain no semantic signal, provided the teacher and student share initialization. The effect persists across data types (numbers, code, chain-of-thought) and model families, but cross-model transfer is limited. A theorem shows a single gradient step can bias the student toward the teacher, raising AI-safety concerns about model provenance and training data.
Scientists report a generative thermodynamic computer that uses thermal noise to produce images from random data, mimicking AI neural networks but with energy use orders of magnitude lower. By leveraging probabilistic computing and diffusion-like dynamics (via Langevin-based calculations) and tuning coupling strengths in a network, the system retrieves or creates images from noise, offering a physics-based path to energy-efficient AI-like tasks and new insights into learning.
Yoshua Bengio, a pioneer in AI and machine learning, has become the first researcher to be cited over one million times on Google Scholar, highlighting his influential work on neural networks and the rapid growth of AI technology.
Research using AI and eye-tracking data reveals that super-recognisers excel at face identification not just by looking more broadly, but by sampling more valuable facial features, which enhances their recognition ability beyond mere quantity of visual information.
Researchers at the University of Surrey have developed a brain-inspired approach called Topographical Sparse Mapping to enhance AI performance and efficiency by mimicking the human brain's neural wiring, reducing energy consumption and improving sustainability in AI models.
Scientists have developed a working computer memory using shiitake mushroom mycelium, demonstrating a low-cost, scalable, and eco-friendly alternative to silicon-based memristors with performance comparable to traditional chips, opening new avenues for biological and neural-inspired computing.
Recent developments and expert opinions suggest that achieving Artificial General Intelligence (AGI) with current Large Language Models (LLMs) is unlikely in the near future, as fundamental challenges like distribution shift remain unresolved and recent AI advancements have fallen short of expectations.
The article argues that in science fiction like Star Trek, computers are depicted as intelligent entities that understand human needs and respond directly without the need for programming or algorithms, suggesting that programming as a craft may become obsolete as AI advances. It highlights how AI in sci-fi is portrayed as autonomous and intuitive, contrasting with current programming practices, and speculates on the future role of programmers.
The article presents a novel enzyme-free DNA circuit architecture that uses heat to reset and recharge molecular systems, enabling sustainable, out-of-equilibrium computation, neural networks, and logic gates without waste build-up, scalable to complex systems like a 100-bit neural network for digit classification.
Researchers have developed neural cellular automata (NCAs) that can reverse-engineer rules to self-assemble into desired shapes and regenerate after damage, mimicking biological processes and offering potential applications in medicine, engineering, and computing.