home.social

#robustness — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #robustness, aggregated by home.social.

  1. New post: AI: The robustness imperative

    If AI is the next infrastructure layer, the real question is not just capability.
    It’s robustness: ownership, open models, digital sovereignty, and resilience under stress.

    I connect Olivier Hamant’s biology lens to AI policy, economics, and commons governance.

    radicaloptimist.org/en/post/ai

    #AI #Robustness #OpenSource #DigitalSovereignty #Commons #PoliticalEconomy

  2. "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
    arxiv.org/abs/2209.15259

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

  3. "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
    arxiv.org/abs/2209.15259

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

  4. "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
    arxiv.org/abs/2209.15259

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

  5. "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
    arxiv.org/abs/2209.15259

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

  6. "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
    arxiv.org/abs/2209.15259

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

  7. The Power Of Using A Story For Better Data Comprehension And Hence Decision Making
    --
    doi.org/10.1080/15228053.2021. <-- 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

  8. The Power Of Using A Story For Better Data Comprehension And Hence Decision Making
    --
    doi.org/10.1080/15228053.2021. <-- 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

  9. 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

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

    jmlr.org/papers/v26/21-0068.ht

    #minimizes #robustness #adversarial

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

    jmlr.org/papers/v26/21-0068.ht

    #minimizes #robustness #adversarial

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

    jmlr.org/papers/v26/21-0068.ht

    #minimizes #robustness #adversarial

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

    jmlr.org/papers/v26/21-0068.ht

    #minimizes #robustness #adversarial

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

    jmlr.org/papers/v26/21-0068.ht

    #minimizes #robustness #adversarial

  15. 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.

    ericmustin.substack.com/p/note

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

  16. 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.

    ericmustin.substack.com/p/note

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

  17. 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.

    ericmustin.substack.com/p/note

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

  18. 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.

    ericmustin.substack.com/p/note

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

  19. 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.

    ericmustin.substack.com/p/note

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

  20. A quotation from Samuel Johnson

    Every quotation contributes something to the stability or enlargement of the language.

    Samuel Johnson (1709-1784) English writer, lexicographer, critic
    A Dictionary of the English Language, Preface (1755)

    Sourcing, notes: wist.info/johnson-samuel/20347…

    #quote #quotes #quotation #English #example #language #nuance #quotation #robustness #dictionary

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

    jmlr.org/papers/v26/22-1428.ht

    #adversarial #regularizing #robustness

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

    jmlr.org/papers/v26/22-1428.ht

    #adversarial #regularizing #robustness

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

    jmlr.org/papers/v26/22-1428.ht

    #adversarial #regularizing #robustness

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

    jmlr.org/papers/v26/22-1428.ht

    #adversarial #regularizing #robustness

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

    jmlr.org/papers/v26/22-1428.ht

    #adversarial #regularizing #robustness

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

    jmlr.org/papers/v25/22-0950.ht

    #adversarial #overfitting #robustness

  27. 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 🥳

    science.org/doi/10.1126/scienc

    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

  28. 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 🥳

    science.org/doi/10.1126/scienc

    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

  29. 'Distributionally Robust Model-Based Offline Reinforcement Learning with Near-Optimal Sample Complexity', by Laixi Shi, Yuejie Chi.

    jmlr.org/papers/v25/22-1482.ht

    #robustness #reinforcement #uncertainty

  30. .@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: plos.io/3K8lIP0 Paper: plos.io/4dXt85p

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

    jmlr.org/papers/v25/21-0233.ht

    #sparse #regularization #robustness