arXiv:2510.20851v1 Announce Type: new
Abstract: Aging is the primary risk factor for nearly all forms of human death, yet pharmaceutical interventions hold the potential to prevent it. Combinations of drugs have been shown to increase the lifespan of model organisms more than individual drugs, and geroprotective histone deacetylase (HDAC) inhibitors that have different molecular targets within the longevity regulation network show considerably higher drug synergy than many other compounds. In this study, four HDAC inhibitors (curcumin, quercetin, resveratrol, and berberine) have been administered in pairwise, three-, and four-combinations to yeast (S. cerevisiae) and their maximum chronological lifespans (CLS) have been measured. In five of the six pairwise combinations, the drugs exhibited synergy according to the Bliss Independence Model, on average extending maximum CLS 68% over the individual drugs. Three- and four-combinations further extended maximum CLS 49% and 107% over pairwise combinations, respectively. Since the targets of the HDAC inhibitors used in this study are evolutionarily conserved between yeast and humans, the results obtained have implications on human longevity.
The Hidden Power of Normalization: Exponential Capacity Control in Deep Neural Networks
arXiv:2511.00958v1 Announce Type: cross Abstract: Normalization methods are fundamental components of modern deep neural networks (DNNs). Empirically, they are known to stabilize optimization dynamics and