Neural Erosion: Emulating Controlled Neurodegeneration and Aging in AI Systems

arXiv:2403.10596v1 Announce Type: new
Abstract: Creating controlled methods to simulate neurodegeneration in artificial intelligence (AI) is crucial for applications that emulate brain function decline and cognitive disorders. We use IQ tests performed by Large Language Models (LLMs) and, more specifically, the LLaMA 2 to introduce the concept of ``neural erosion." This deliberate erosion involves ablating synapses or neurons, or adding Gaussian noise during or after training, resulting in a controlled progressive decline in the LLMs' performance. We are able to …

abstract aging ai systems applications artificial artificial intelligence arxiv brain cognitive concept cs.ai cs.cl function intelligence language language models large language large language models llama llama 2 llms q-bio.nc systems tests type