Operating on a minimal energy budget, the human brain is able to efficiently process vast amounts of temporal information at different timescales as it quickly learns to act in new environments. By contrast, current AI models do not learn temporal information efficiently, struggle with lifelong learning – the ability to keep learning new tasks continuously throughout life – and also do not perform well in resource-constrained environments. This project aims to create new AI models that overcome these limitations by leveraging mechanisms inspired by theories for how the brain is able to efficiently learn temporal information. Particularly, the project is based on a recent theory, Temporal Scaffolding, which postulates that during sleep, the brain reactivates wake experiences in an accelerated manner to allow detecting important temporal patterns embedded in those experiences. The goal of this project is to develop autonomous machines, informed by the temporal scaffolding hypothesis, which can rapidly adapt, operate under uncertainty, and evolve throughout their lifespan despite resource constraints. This transformative approach has the potential to address major AI challenges and find applications in healthcare, energy, and national security. The team aims to promote broad access to the computational strategies through initiatives at multiple educational institutions, emphasizing cross-disciplinary training and outreach to underrepresented populations. The team will conduct value-sensitive workshops and regular ethics consultations throughout the project. Alongside the technical goals, the team aims to offer opportunities for underrepresented students in AI fields, fostering a competitive AI workforce to maintain US technological leadership in STEM. By emulating how the human brain learns, the team seeks to create efficient, lifelong learning AI systems capable of revolutionizing various industries and benefiting society as a whole.
The Temporal Scaffolding Hypothesis provides a novel explanation for the brain’s superior ability to efficiently learn temporal information. According to this hypothesis, time-compressed memory replay during offline periods serves to extract temporal regularities within encoded experiences. Building on the temporal scaffolding hypothesis, in the present project the PIs propose a set of mechanisms underlying resource-efficient lifelong learning of spatiotemporal regularities employing online (wake) and offline (sleep) periods, which they intend to both verify in new human experiments and incorporate in machine learning algorithms. Advances in theory, models, and systems stemming from this grant will have applications in multiple domains.
The two specific aims for this project are to:
- Develop new AI algorithms and architectures, inspired by the temporal scaffolding hypothesis, for efficient learning of spatio-temporal patterns
- Extend the temporal scaffolding hypothesis to include hierarchical representations that support lifelong learning and verify the predictions of the model through human experiments and computational investigations.
Through these aims the researchers will develop optimization frameworks that support deployment in resource constrained environments. Moreover, this project will yield scalable deep neural network and spiking neural network models that incorporate temporally compressed replay mechanisms. This approach is expected to limit the catastrophic interference effects that hinder most current network models of memory and improve the system’s capacity for lifelong learning. Training and access to these transformative computational strategies will be broadened via multiple initiatives at the University of Texas at San Antonio, the University of Rochester, and the University of Tennessee, Knoxville, including successful K-12 partnerships and targeted experiential outreach strategies. The team will also engage in ethical design through value-sensitive workshops and regular ethics consultations. The project design efforts will also provide significant opportunities to underrepresented students in cross-cutting AI fields and promote a robust and competitive AI workforce that maintains US technological leadership in STEM.
Core Members
Publications
Book Chapters
- Book Chapter on continual learning – van de Ven, Gido M., Nicholas Soures, and Dhireesha Kudithipudi. “Continual Learning and Catastrophic Forgetting.” arXiv preprint arXiv:2403.05175 (2024).
Journals or Juried Conference Paper
- Zohora, Fatima Tuz, et al. “Probabilistic metaplasticity for continual learning with memristors in spiking networks”. Under review in Scientific Reports. NSF EFRI BRAID # 2317706 Efficient Learning of Spatiotemporal Regularities in Humans and Machines through Temporal Scaffolding.
- Dabool, Alashwal, Lerner & Moustafa. “Overcoming Catastrophic Interference: Strategies for Continuous Learning in Neural Networks”. [In preparation]
Other Conference Presentation or Paper
- Patel, Raghav, Nicholas Soures, and Dhireesha Kudithipudi. “Does Replay Suffice for Online Continual Learning in Spiking Networks?” Cognitive Computational Neuroscience 2024.
- Karia, Vedant, Abdullah Zyarah, and Dhireesha Kudithipudi. “PositCL: Compact Continual Learning with Posit Aware Quantization.” Proceedings of the Great Lakes Symposium on VLSI 2024
- Soures, Nicholas, Vedant Karia, and Dhireesha Kudithipudi. “Advancing Neuro-Inspired Lifelong Learning for Edge with Co-Design.” Proceedings of the AAAI Symposium Series. Vol. 3. No. 1. 2024.
- Zyarah, Abdullah M., and Dhireesha Kudithipudi. “Time-Series Forecasting and Sequence Learning Using Memristor-based Reservoir System.” arXiv preprint arXiv:2405.13347 (2024).
- Hickok, Truman, and Dhireesha Kudithipudi. “Watch Your Step: Optimal Retrieval for Continual Learning at Scale.” arXiv preprint arXiv:2404.10758 (2024).
- Zohora, Fatima Tuz, et al. “Probabilistic Metaplasticity for Continual Learning with Memristors.” arXiv preprint arXiv:2403.08718 (2024).
- Harun, Yousuf Md, Lee, Kyungbok, Gallardo, Jhair, Krishnan, Giri, Kanan, Christopher “What variables affect out-of-distribution generalization in pretrained models?” arXiv:2405.15018 (2024).
- Harun, Yousuf Md, Kanan, Christopher “Overcoming the Stability Gap in Continual Learning.” arxiv:2306.01904 (2024).