A Caltech researcher presented at MBZUAI on memory representation and retrieval, contrasting AI and neuroscience approaches. Current AI retrieval systems like RAG retrieve via fine-tuning and embedding similarity, while the presenter argued for exploring retrieval via combinatorial object identity or spatial proximity. The research explores circuit-level retrieval via domain fine-tuned LLMs and distributed memory for image retrieval using semantic similarity. Why it matters: The work suggests structured databases and retrieval-focused training can allow smaller models to outperform larger general-purpose models, offering efficiency gains for AI development in the region.
Tom M. Mitchell from Carnegie Mellon University discussed using machine learning to study how the brain processes natural language, using fMRI and MEG to record brain activity while reading text. The research explores neural encodings of word meaning, information flow during word comprehension, and how meanings of words combine in sentences and stories. He also touched on how understanding of the brain aligns with current AI approaches to NLP. Why it matters: This interdisciplinary research could bridge the gap between neuroscience and AI, potentially leading to more human-like NLP models.
An article in Forbes highlights the Mohammed bin Zayed University of Artificial Intelligence (MBZUAI) as the first university devoted exclusively to AI advancement. MBZUAI President Eric Xing champions a 'connectionism' approach, designing computational models inspired by interconnected human cognition networks. AI's ability to process and analyze data at high speeds unlocks new knowledge realms, acting as a universal translator between humans and the digital world. Why it matters: MBZUAI is positioned as a key institution driving AI innovation and responsible AI practices in the Middle East.
MBZUAI researchers are developing spiking neural networks (SNNs) to emulate the energy efficiency of the human brain. Traditional deep learning models like those powering ChatGPT consume significant energy, with a single query using 3.96 watts. SNNs aim to mimic biological neurons more closely to reduce energy consumption, as the human brain uses only a fraction of the energy compared to these models. Why it matters: This research could lead to more sustainable and energy-efficient AI technologies, addressing a major challenge in deploying large-scale AI systems.
Prof. Mérouane Debbah of the Technology Innovation Institute (TII) warns that current AI development relies on unsustainable, energy-intensive "bruteforce computing." He argues that the field needs more energy-efficient algorithms instead of simply scaling up GPUs. Debbah suggests neuromorphic computing as a potential solution, drawing inspiration from the human brain's energy efficiency. Why it matters: This critique highlights a crucial sustainability challenge for AI in the GCC and globally, as the region invests heavily in compute-intensive AI models.
Liangming Pan from UCSB presented research on building reliable generative AI agents by integrating symbolic representations with LLMs. The neuro-symbolic strategy combines the flexibility of language models with precise knowledge representation and verifiable reasoning. The work covers Logic-LM, ProgramFC, and learning from automated feedback, aiming to address LLM limitations in complex reasoning tasks. Why it matters: Improving the reliability of LLMs is crucial for high-stakes applications in finance, medicine, and law within the region and globally.