#robustness — Public Fediverse posts

"We then survey statistical lower bounds that, we argue, constitute a compelling case against the possibility of designing high-accuracy LAIMs with strong security guarantees."

On the Impossible Safety of Large AI Models
https://arxiv.org/abs/2209.15259

#generativeAI #chatBots #LLMs #safety #personalSafety #robustness #genAI #accuracy

The Power Of Using A Story For Better Data Comprehension And Hence Decision Making
--
https://doi.org/10.1080/15228053.2021.2016151 <-- shared book review, “Data Story: Explain Data And Inspire Action Through Story”
--
[I encountered this excellent graphic from @saurabh Rai, and went and explored the ideas put so succinctly here; I found, well, a technical story overview (link above) to ‘match’; however, this should not be considered an endorsement of this book]
#data #storytelling #data #comprehension #presentation #story #frameworks #context #setting #dataquality #communication #usecase #robustness #insights #correctness #decisionmaking #narratives #decisions

Evaluating The Functional Realism Of Deep Learning Rainfall-Runoff Models Using Catchment Hydrology Principles
--
https://doi.org/10.1029/2025WR040076 <-- shared paper
--
#water #hydrology #surfacewater #pluvial #fluvial #rainfall #snow #snowmelt #runoff #precipitation #model #blackbox #robustness #functionalrealism #screening #parameters #accuracy #hydrologic #principles #trustworthy #modeling #spatialanalysis #spatial #mapping #GIS #spatiotemporal #USA #CONUS #AI #ExplainableAI #celerity #machinelearning #artificialintelligence #LSTM #deeplearning #evapotranspiration #waterresources #extremeweather #flood #flooding #risk #hazard #monitoring #prediction #catchments #streamflow #geomorphometry #network #flow #calibration

Evaluating The Functional Realism Of Deep Learning Rainfall-Runoff Models Using Catchment Hydrology Principles
--
https://doi.org/10.1029/2025WR040076 <-- shared paper
--
#water #hydrology #surfacewater #pluvial #fluvial #rainfall #snow #snowmelt #runoff #precipitation #model #blackbox #robustness #functionalrealism #screening #parameters #accuracy #hydrologic #principles #trustworthy #modeling #spatialanalysis #spatial #mapping #GIS #spatiotemporal #USA #CONUS #AI #ExplainableAI #celerity #machinelearning #artificialintelligence #LSTM #deeplearning #evapotranspiration #waterresources #extremeweather #flood #flooding #risk #hazard #monitoring #prediction #catchments #streamflow #geomorphometry #network #flow #calibration

Evaluating The Functional Realism Of Deep Learning Rainfall-Runoff Models Using Catchment Hydrology Principles
--
https://doi.org/10.1029/2025WR040076 <-- shared paper
--
#water #hydrology #surfacewater #pluvial #fluvial #rainfall #snow #snowmelt #runoff #precipitation #model #blackbox #robustness #functionalrealism #screening #parameters #accuracy #hydrologic #principles #trustworthy #modeling #spatialanalysis #spatial #mapping #GIS #spatiotemporal #USA #CONUS #AI #ExplainableAI #celerity #machinelearning #artificialintelligence #LSTM #deeplearning #evapotranspiration #waterresources #extremeweather #flood #flooding #risk #hazard #monitoring #prediction #catchments #streamflow #geomorphometry #network #flow #calibration

Evaluating The Functional Realism Of Deep Learning Rainfall-Runoff Models Using Catchment Hydrology Principles
--
https://doi.org/10.1029/2025WR040076 <-- shared paper
--
#water #hydrology #surfacewater #pluvial #fluvial #rainfall #snow #snowmelt #runoff #precipitation #model #blackbox #robustness #functionalrealism #screening #parameters #accuracy #hydrologic #principles #trustworthy #modeling #spatialanalysis #spatial #mapping #GIS #spatiotemporal #USA #CONUS #AI #ExplainableAI #celerity #machinelearning #artificialintelligence #LSTM #deeplearning #evapotranspiration #waterresources #extremeweather #flood #flooding #risk #hazard #monitoring #prediction #catchments #streamflow #geomorphometry #network #flow #calibration

Evaluating The Functional Realism Of Deep Learning Rainfall-Runoff Models Using Catchment Hydrology Principles
--
https://doi.org/10.1029/2025WR040076 <-- shared paper
--
#water #hydrology #surfacewater #pluvial #fluvial #rainfall #snow #snowmelt #runoff #precipitation #model #blackbox #robustness #functionalrealism #screening #parameters #accuracy #hydrologic #principles #trustworthy #modeling #spatialanalysis #spatial #mapping #GIS #spatiotemporal #USA #CONUS #AI #ExplainableAI #celerity #machinelearning #artificialintelligence #LSTM #deeplearning #evapotranspiration #waterresources #extremeweather #flood #flooding #risk #hazard #monitoring #prediction #catchments #streamflow #geomorphometry #network #flow #calibration

Implicit Ode Solvers Are Not Universally More Robust Than Explicit Ode Solvers

https://www.stochasticlifestyle.com/implicit-ode-solvers-are-not-universally-more-robust-than-explicit-ode-solvers-or-why-no-ode-solver-is-best/

#HackerNews #ImplicitOdeSolvers #ExplicitOdeSolvers #NumericalAnalysis #ComputationalMath #Robustness #Algorithms

Hey FOSS-community, I'm trying to compare #GitLab and @forgejo.

Some important aspects are:

- #DigitalSovereignty (most important)
- #SelfHosting
- #Collaboration between instances (#federation)
- #PublicValues
- Prevention of #VendorLockIn
- #Scalability
- #Robustness
- Guaranteed #support by professionals
- Software #freedom
- #Longevity
- Suitability for the #government

Do you have any thoughts on this? Or any useful links?

cc: @Codeberg

#Git #GitForge #opensource #FOSS #community

'Towards Optimal Branching of Linear and Semidefinite Relaxations for Neural Network Robustness Certification', by Brendon G. Anderson, Ziye Ma, Jingqi Li, Somayeh Sojoudi.

http://jmlr.org/papers/v26/21-0068.html

#minimizes #robustness #adversarial

I really liked this informal community poll and thematic analysis on SLO usage. It does a better job at highlighting the hurdles to adopting them at a Company Who Is Not Google than a lot of "Here's how to do SLOs" pieces that just don't cover it.

If there is ever a "Seeking SLOs" book, this should be the first chapter.

https://ericmustin.substack.com/p/notes-on-service-level-objectives

#SRE #SLO #Reliability #Robustness #ErrorBudgets #SLI #SLA

A more robust raw OpenBSD syscall demo

https://nullprogram.com/blog/2025/03/06/

#HackerNews #OpenBSD #Syscalls #Demo #Robustness #Technology #Innovation

'Regularizing Hard Examples Improves Adversarial Robustness', by Hyungyu Lee, Saehyung Lee, Ho Bae, Sungroh Yoon.

http://jmlr.org/papers/v26/22-1428.html

#adversarial #regularizing #robustness

'On the Impact of Hard Adversarial Instances on Overfitting in Adversarial Training', by Chen Liu, Zhichao Huang, Mathieu Salzmann, Tong Zhang, Sabine Süsstrunk.

http://jmlr.org/papers/v25/22-0950.html

#adversarial #overfitting #robustness

'Label Noise Robustness of Conformal Prediction', by Bat-Sheva Einbinder et al.

http://jmlr.org/papers/v25/23-1549.html

#adversarial #robustness #prediction

Centuries ago, I published a manuscript on BioRχiv about the love/hate relationship between variability and robustness in embryogenesis.

And now, it's published in Science! I am so proud 🥳

https://www.science.org/doi/10.1126/science.adh1145

We went through three (3!) rounds of revisions. Although we did not change a comma in the abstract, the revisions improved the quality of the work very much (we nearly doubled the sample size and added an entire figure).

#variability #robustness #embryology #morphogenesis #mouse

Centuries ago, I published a manuscript on BioRχiv about the love/hate relationship between variability and robustness in embryogenesis.

And now, it's published in Science! I am so proud 🥳

https://www.science.org/doi/10.1126/science.adh1145

We went through three (3!) rounds of revisions. Although we did not change a comma in the abstract, the revisions improved the quality of the work very much (we nearly doubled the sample size and added an entire figure).

#variability #robustness #embryology #morphogenesis #mouse

.@bphmetz explores a #PLOSBiology study which uses in silico rewiring of #GeneticCodes & functional protein data to reveal a positive correlation between #robustness & protein #evolvability. Primer: https://plos.io/3K8lIP0 Paper: https://plos.io/4dXt85p

'Sparse NMF with Archetypal Regularization: Computational and Robustness Properties', by Kayhan Behdin, Rahul Mazumder.

http://jmlr.org/papers/v25/21-0233.html

#sparse #regularization #robustness

Updated 4th Chapter of my "#EvolutionarySystems and #BioInspiredComputing " 2024 Class. #SelfOrganization #Criticality #Robustness #Resilience #Evolvability
https://casci.binghamton.edu/academics/i-bic/lec04.php

Robustness and resilience of Complex Networks: the new essay in the last #ComplexityThoughts featuring also Shlomo Havlin, Alex Arenas, @PessoaBrain , Giulia Menichetti, Marta Gonzàlez to discuss applications from cells to social & urban systems.

#ComplexSystems #ComplexNetworks #Robustness #Resilience

Don't miss it!

https://open.substack.com/pub/manlius/p/robustness-and-resilience-in-complex?r=n4iyo&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true

The influential American number theorist Leonard Dickson wrote 'Thank God that number theory is unsullied by any application.'

And yet, again and again, number theory finds unexpected applications in science and engineering, from leaf angles that (almost) universally follow the Fibonacci sequence, to modern encryption techniques based on factoring prime numbers.

Now, researchers have demonstrated an unexpected link between number theory and evolutionary genetics.

Specifically, the team of researchers - from Oxford, Harvard, Cambridge, GUST, MIT, Imperial, and the Alan Turing Institute - have discovered a deep connection between the sums-of-digits function of number theory and a key quantity in genetics: #phenotype #mutational #robustness.

This quality is defined as the average probability that a point mutation does not change a phenotype (a characteristic of an organism).

They proved that the maximum robustness is proportional to the logarithm of the fraction of all possible sequences that map to a phenotype, with a correction which is given by the sums of digits function sk(n), defined as the sum of the digits of a natural number n in base k. For example, for n = 123 in base 10, the digit sum would be s10(123) = 1 + 2 + 3 = 6.

Another surprise was that the maximum robustness also turns out to be related to the famous Tagaki function, a bizarre function that is continuous everywhere, but differentiable nowhere.

First author Dr. Vaibhav Mohanty (Harvard Medical School) added: 'What is most surprising is that we found clear evidence in the mapping from sequences to RNA secondary structures that nature in some cases achieves the exact maximum robustness bound. It's as if biology knows about the fractal sums-of-digits function.'
https://www.news-medical.net/news/20230808/Unexpected-link-between-pure-mathematics-and-genetics-discovered.aspx

Vulnerability-Aware Instance Reweighting For Adversarial Training

Olukorede Fakorede, Ashutosh Kumar Nirala, Modeste Atsague, Jin Tian

Action editor: Qibin Zhao.

https://openreview.net/forum?id=kdPcLdJbt1

#adversarial #classifiers #robustness

4/

Below, key points:

- "lack of #ModelEvaluation"

- statistical #uncertainty & "#robustness of event attribution results"

#References

[4] Seneviratne, et al., 2021. Chapter 11: weather and climate extreme events in a changing climate. In: Climate Change 2021: The Physical Science Basis - Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. IPCC, Geneva, Switzerland, pp. 1513–1766. https://purl.org/INRMM-MiD/z-ED8RQFV5

https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6_WGI_Chapter11.pdf#page=28

🔍 #Scan your #AI model to detect potential vulnerabilities prior to deployment, including #performance, #spurious correlations, data leakage, non-#robustness, ethical #biases, overconfidence, and more. [2/7]

@AxelPolleres Thank you very much for the link!
They have a live video stream:
https://caiml.dbai.tuwien.ac.at/news/47/#live-stream

Currently watching Wolfgang Nejdl talking about "On the Role of Causality for Trustworthy Artificial Intelligence"

Panel discussion up next with Antonio A. Casilli, Gerti Kappel, Wolfgang Nejdl, Helga Nowotny, John Shawe-Taylor

YouTube: https://www.youtube.com/watch?v=SSTabpCj0QM

#ai #digitalhumanism #artificialintelligence #vienna #caiml2023 #interpretability #fairness #robustness #privacy

Is Economics Self-Correcting?
Replications in the American Economic Review
https://www.rwi-essen.de/fileadmin/user_upload/RWI/Publikationen/Ruhr_Economic_Papers/REP_23_1005.pdf
“…most comments are barely cited, & they have no impact on the original papers’ subsequent citations.
…#replications do not update the #economics literature
…conventional wisdom that #robustness & #replicability are hard to define in economics.”

Is Economics Self-Correcting?
Replications in the American Economic Review
https://www.rwi-essen.de/fileadmin/user_upload/RWI/Publikationen/Ruhr_Economic_Papers/REP_23_1005.pdf
“…most comments are barely cited, & they have no impact on the original papers’ subsequent citations.
…#replications do not update the #economics literature
…conventional wisdom that #robustness & #replicability are hard to define in economics.”

Is Economics Self-Correcting?
Replications in the American Economic Review
https://www.rwi-essen.de/fileadmin/user_upload/RWI/Publikationen/Ruhr_Economic_Papers/REP_23_1005.pdf
“…most comments are barely cited, & they have no impact on the original papers’ subsequent citations.
…#replications do not update the #economics literature
…conventional wisdom that #robustness & #replicability are hard to define in economics.”

Is Economics Self-Correcting?
Replications in the American Economic Review
https://www.rwi-essen.de/fileadmin/user_upload/RWI/Publikationen/Ruhr_Economic_Papers/REP_23_1005.pdf
“…most comments are barely cited, & they have no impact on the original papers’ subsequent citations.
…#replications do not update the #economics literature
…conventional wisdom that #robustness & #replicability are hard to define in economics.”

Is Economics Self-Correcting?
Replications in the American Economic Review
https://www.rwi-essen.de/fileadmin/user_upload/RWI/Publikationen/Ruhr_Economic_Papers/REP_23_1005.pdf
“…most comments are barely cited, & they have no impact on the original papers’ subsequent citations.
…#replications do not update the #economics literature
…conventional wisdom that #robustness & #replicability are hard to define in economics.”

The Fundamental Limits of Neural Networks for Interval Certified Robustness

Matthew B Mirman, Maximilian Baader, Martin Vechev

https://openreview.net/forum?id=fsacLLU35V

#interval #robustness #robust

#Introduction

Hi all 👋, I am a first year PhD student in #NLProc at the University of Mannheim working on #ScholarlyNLP / #ScholarlyDocumentProcessing and #ComputationalSocialScience.

🔍 My work involves identifying survey variable mentions in scientific texts #svident

💡 I am also interested in #InformationRetrieval, #EntityLinking, #LegalNLP, #Explainability, #Fairness, and #Robustness.

#Introduction again now that I have migrated to the evoeco instance. Some hashtags that align with my interests: #science
#evolution #popgen #biology #theory #proteins #bioinformatics #datascience #aging #prions #evolvability #robustness #astrobiology #epidemiology #COVID19 #economics #RiskAnalysis

Belated #introduction with some hashtags that align with my interests: #science
#evolution #popgen #biology #theory #proteins #bioinformatics #datascience #aging #prions #evolvability #robustness #astrobiology #epidemiology #COVID19 #economics #RiskAnalysis