Netholabs generates large-scale neural-behavioral datasets using advanced mouse neural recording systems and ethology-focused behavior arenas. This data is used to create whole-brain functional dynamics models, such as whole-brain emulators (WBEs), for neuroscience research.
Funding
Funding not disclosed
Founders
Product
Problem
Traditional neuroscience research relies on single-task, low-bandwidth behavior assays, creating bottlenecks in data collection and limiting the understanding of naturalistic behaviors. Analyzing neural circuits involved in brain disorders is hampered by manual workflows and siloed data, hindering translational research and replication efforts.
Solution
NethoLabs develops advanced AI-driven neuroethology platforms that generate large-scale, multimodal neural-behavioral datasets. These platforms utilize naturalized behavior rodent habitats and advanced neural recording systems to capture long-term, high-resolution behavioral data. The resulting datasets are used to build whole-brain functional dynamics models, including foundation models (FMs) and whole-brain emulators (WBEs), which emulate mammalian brain dynamics based on causal neuro-behavior models. These models provide a novel perspective into neural function in health and disease from a whole-organism perspective and offer a transparent substrate for AI alignment.
Target Audience
NethoLabs serves neuroscientists, AI researchers, and organizations focused on AI safety, particularly those interested in advanced neuroscience research platforms, whole-brain emulation, and understanding the neural basis of behavior.
Features
- AI-driven neuroethology platforms for collecting and analyzing data over extended periods.
- Naturalized behavior rodent habitats for high-resolution, long-term behavioral recording.
- Generation of petabyte-scale multimodal datasets of neural and behavioral activity.
- Development of foundation models (FMs) and whole-brain emulators (WBEs) for simulating brain dynamics.
- Tools for analyzing and decomposing models using neuroscience and AI interpretability techniques.
- Scalable hardware and data generation approaches with AI acceleration.