Nvidia’s new chips target the gaming, robotics, and autonomous vehicles industries, partnerships with Toyota and others highlight Nvidia’s AI expansion in self-driving technology.
Neuromodulators in the brain act globally at many forms of synaptic plasticity, represented as metaplasticity, which is rarely considered by existing spiking (SNNs) and nonspiking artificial neural networks (ANNs). Here, we report an efficient brain-inspired computing algorithm for SNNs and ANNs, referred to here as neuromodulation-assisted credit assignment (NACA), which uses expectation signals to induce defined levels of neuromodulators to selective synapses, whereby the long-term synaptic potentiation and depression are modified in a nonlinear manner depending on the neuromodulator level. The NACA algorithm achieved high recognition accuracy with substantially reduced computational cost in learning spatial and temporal classification tasks. Notably, NACA was also verified as efficient for learning five different class continuous learning tasks with varying degrees of complexity, exhibiting a markedly mitigated catastrophic forgetting at low computational cost. Mapping synaptic weight changes showed that these benefits could be explained by the sparse and targeted synaptic modifications attributed to expectation-based global neuromodulation.
Using an analysis of a voluntary action caused by a visual perception, I suggest that the three fundamental characteristics of this perception (being conscious, self-conscious, and provided with a content) are neurologically implemented by three distinct higher order properties of brain dynamics. This hypothesis allows me to sketch out a physicalist naturalist solution to the mind-body problem. According to this solution, primary phenomenal consciousness is neither a non-physical substance, nor a non-physical property but simply the “format” that the brain gives to a part of its dynamics in order to obtain a fine tuning with its environment when the body acts on it.
The relationship between brain and computer is a perennial theme in theoretical neuroscience, but it has received relatively little attention in the philosophy of neuroscience. This paper argues that much of the popularity of the brain-computer comparison (e.g. circuit models of neurons and brain areas since McCulloch and Pitts, Bull Math Biophys 5: 115–33, 1943) can be explained by their utility as ways of simplifying the brain. More specifically, by justifying a sharp distinction between aspects of neural anatomy and physiology that serve information-processing, and those that are ‘mere metabolic support,’ the computational framework provides a means of abstracting away from the complexities of cellular neurobiology, as those details come to be classified as irrelevant to the (computational) functions of the system.
CMOS manufacturing has been harnessed for optical transceivers, and some companies are now looking to build photonic circuits for sensing, quantum science and more.
Researchers have demonstrated a new method for storing and retrieving quantum information, marking a leap forward for quantum memory.
Researchers have discovered that certain disordered superconductors exhibit abrupt phase transitions, a finding that challenges established theories and could have implications for quantum computing.
A study published in Nature by researchers investigating indium oxide films — a highly disordered superconductor — shows that their transition from a superconducting to an insulating state is not gradual, as traditionally assumed, but sudden. This abrupt shift, known as a first-order quantum phase transition, contrasts with the commonly observed continuous, second-order transitions in superconductors.
Key measurements revealed a sharp drop in superfluid stiffness — which is a property that reflects the superconducting state’s ability to resist phase distortions — at a critical level of disorder. Interestingly, the critical temperature of these films, where superconductivity breaks down, no longer depended on the strength of electron pairing but rather on the superfluid stiffness. This behavior aligns with a pseudogap regime, where electron pairs exist but lack the coherence needed for superconductivity.
Join NVIDIA CEO Jensen Huang at CES on January 6 at 6:30 p.m. PT for an exclusive look at groundbreaking innovations in AI, gaming, robotics, and beyond.
Google is establishing a team to build AI that can simulate the physical world, led by one of the former co-leads on OpenAI’s Sora model.
