Tag
The paper proposes MAPA, a self-supervised pretraining method with spatial encodings for neural interfaces, which reduces the labeled data needed for accurate brain-computer interface decoding across subjects.
EEG-VID introduces task-guided latent predictive pretraining to improve EEG decoding under session and subject shifts, achieving above-chance target selection in assistive robotics scenarios.
Neuralink enters mass production of its brain implant, but a 1974 patent for remote brain wave monitoring reveals the concept is not new, underscoring decades of neural interface research.
This paper introduces Differentiable Logic Gate Networks (Diff-Logic) as a hardware-native alternative to conventional neural networks for real-time EEG classification on edge devices, achieving competitive performance with significantly lower latency and model size.
OpenAI announces investment in Merge Labs' seed round, a research lab developing brain-computer interfaces (BCIs) that combine biology, devices, and AI to enable high-bandwidth human-AI interaction. OpenAI will collaborate with Merge Labs on scientific foundation models to accelerate BCI research and development.
The article discusses the current state and future of brain-computer interfaces (BCIs), highlighting Science's Prima retinal implant that has restored partial vision in over 40 patients. It explores neural plasticity, the trade-off between surgical risks and benefits, and the potential for BCIs to eventually enhance healthy individuals.