Download full-text PDF |
Source |
|---|
Publication Analysis
Top Keywords
Similar Publications
Replay as a Basis for Backpropagation Through Time in the Brain.
Neural Comput
January 2025
Department of Psychological and Brain Sciences, Indiana University Bloomington, Bloomington, IN 47405, U.S.A.
How episodic memories are formed in the brain is a continuing puzzle for the neuroscience community. The brain areas that are critical for episodic learning (e.g.
View Article and Find Full Text PDFEnvironmental interactions between protists and bacterial communities in hydrocarbon degradation.
Protist
December 2024
C. G. Bhakta Institute of Biotechnology, Uka Tarsadia University, Maliba Campus, Bardoli, Surat 394 350, Gujarat, India. Electronic address:
Reclamation of petroleum-polluted environments is a key issue for today and in the future, as our reliance on oil will persist for decades. An eco-friendly solution is to use microbes that play a role in petroleum-hydrocarbon degradation. However, as hydrocarbon degradation involves a multi-step process involving different functional groups, focusing only on finding efficient bacterial species will not be the complete solution.
View Article and Find Full Text PDFSolving the robustness puzzle: The joint impact of optimization approach, robustness metrics, and scenarios on water resources management under deep uncertainty.
J Environ Manage
January 2025
Department of Electronics, Information, and Bioengineering, Politecnico di Milano, Milano, I-20133, Italy.
The concept of robustness has been widely used in water resources management to identify solutions that perform satisfactorily across a range of plausible future conditions to increase confidence in decision-making in a deeply uncertain future. However, the selection of an appropriate metric to quantify robustness remains challenging due to the existence of multiple choices reflecting different risk preferences. In addition, different scenarios can be used to represent plausible future conditions, which adds another layer of complexity to solution identification.
View Article and Find Full Text PDFHow Do Transformers Model Physics? Investigating the Simple Harmonic Oscillator.
Entropy (Basel)
November 2024
Institute for Artificial Intelligence and Fundamental Interactions, Massachusetts Institute of Technology, Cambridge, MA 02139, USA.
How do transformers model physics? Do transformers model systems with interpretable analytical solutions or do they create an "alien physics" that is difficult for humans to decipher? We have taken a step towards demystifying this larger puzzle by investigating the simple harmonic oscillator (SHO), x¨+2γx˙+ω02x=0, one of the most fundamental systems in physics. Our goal was to identify the methods transformers use to model the SHO, and to do so we hypothesized and evaluated possible methods by analyzing the encoding of these methods' intermediates. We developed four criteria for the use of a method within the simple test bed of linear regression, where our method was y=wx and our intermediate was : (1) Can the intermediate be predicted from hidden states? (2) Is the intermediate's encoding quality correlated with the model performance? (3) Can the majority of variance in hidden states be explained by the intermediate? (4) Can we intervene on hidden states to produce predictable outcomes? Armed with these two correlational (1,2), weak causal (3), and strong causal (4) criteria, we determined that transformers use known numerical methods to model the trajectories of the simple harmonic oscillator, specifically, the matrix exponential method.
View Article and Find Full Text PDFModels of Fluctuating Selection Between Generations: A Solution for the Theoretical Inconsistency.
J Mol Evol
December 2024
Department of Genetics, Development and Cell Biology, Iowa State University, Ames, Iowa, IA, 50011, USA.
Article Synopsis
- The theory of selection fluctuation between generations has been widely discussed in population genetics since the 1970s, highlighting that mutation frequencies may oscillate around 0.5 before becoming fixed or lost.
- The concept of a "reference phenotype" is introduced, which posits that one of the three genotypes (AA, AA', A'A') must retain constant fitness to support fluctuating selection.
- A proposed mixture model suggests that all three genotypes should be considered as potential reference types, revealing that balancing selection emerges based on their relative frequencies.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!