ANI: Adaptive Neural Intelligence
Neural Sensing, Adaptive Representation, Edge NeuroAI, and Extreme Environments
Introduction
The convergence of biological neural systems, artificial intelligence, advanced sensing, and edge computing is opening a new frontier in human-machine intelligence. Brain-Computer Interfaces (BCIs), high-density EEG, fNIRS, wearable neurotechnology, and multimodal sensing are moving beyond controlled laboratories toward clinical, consumer, robotic, and extreme environments.
This session focuses on the frontier of neural intelligence across this continuum. It begins with robust neural sensing and structured representations of noisy, non-stationary signals, advances toward adaptive representation that can generalize across users, sessions, and environments, and brings these capabilities to edge intelligence for resource-constrained, real-time inference. These foundations enable reliable human-machine systems and ultimately extend neural intelligence into extreme environments, including future space missions.
This progression connects machine learning, neuroscience, neuroengineering, robotics, embedded systems, and aerospace technology around fundamental advances in geometric and statistical neural representations, dynamic covariance modeling, self-supervised learning, continual adaptation, zero-calibration decoding, NeuroAI-hardware co-design, and efficient neural computing.
The session aims to advance neural intelligence from static laboratory decoding toward adaptive, resource-aware, and robust systems capable of operating across users, sensors, environments, and missions.
Topics
Adaptive Neural Learning
- Riemannian geometry, Symmetric Positive Definite (SPD) learning, and covariance estimation
- Dynamic neural-state and covariance-trajectory modeling
- Self-supervised and contrastive neural representation learning
- Continual learning and test-time adaptation
- Zero-/few-calibration and cross-subject neural decoding
- Domain generalization and robust neural representations
NeuroAI and Edge Systems
- TinyML and ultra-low-power neural decoding
- Quantization, pruning, and knowledge distillation
- Hardware-aware neural representation learning
- Adaptive sensor selection and multimodal fusion
- EEG-fNIRS, ear-EEG, wearable, and novel neural sensors
- Real-time closed-loop neural control
Extreme and Space Environments
- Astronaut cognitive-state and workload monitoring
- Neural decoding under altered physiology and motion
- Robust neural sensing under resource constraints
- Human-machine teaming in extreme environments
- Neural interfaces for autonomous systems and space habitats
- Terrestrial analogues and validation methodologies for space deployment
Safety, Privacy, and Responsible Neurotechnology
- Privacy-preserving and federated neural learning
- Neural-data security and governance
- Fairness and cross-population robustness
- Uncertainty estimation and safe adaptive neural systems
- Clinical and operational validation
ANI particularly encourages contributions combining new theory, algorithms, datasets, hardware, and reproducible experimental systems.
Call for Submissions
We welcome original research and practical work on the topics above in three forms:
- Papers: Full, Short, Position/Vision, Poster, and Demo Papers.
- Abstracts: Poster-Only, Demo-Only, Lightning Talk, and Previously Published Work.
- Panel Proposals: Moderated panels of three to five speakers from diverse fields, sectors, or perspectives.
Position and vision contributions are especially welcome, as Position/Vision Papers or as Lightning Talk abstracts.
Join a panel. Cross-AI hosts panels on open questions that span the Frontier Sessions and workshops. Apply as a panelist or moderator through the Call for Panelists.
Submission Guidelines
Submissions are made through the AirBalloon conference management system. All papers are reviewed double-blind.
Before submitting, read the Author Guidelines, Submission Types, the Submission Guidelines, the Publication and Media Hosting guidelines, and all policies, then complete the Checklist for Initial Submission. Submissions that do not follow the guidelines and policies are desk-rejected under the Desk Rejection Policy.
Call for Reviewers
To serve on the workshop program committee or as a reviewer, please apply online. Applicants may select multiple workshops relevant to their expertise. Reviewers must have an academic, industry, or other organizational affiliation; independent applicants are not considered. If you do not have an AirBalloon account, register using your institutional or organizational work email address. Personal and professional society email addresses (e.g., ieee.org or acm.org) are not accepted. If you already have an account, ensure your email address is a qualifying work email and that your bio, affiliation, and website link are complete. Incomplete profiles will not be considered. Approved applicants may be invited to review papers. Qualified reviewers will be acknowledged as external reviewers. Those providing high-quality reviews will be invited to join the Program Committee.