#backwriters - Twitter Hashtag

1 day ago

#BlueWave #Democrats #MenCan #BoysCan #Empathy #RightToGuardHumanLife #ErelimSec #SeraphimSec #HelpFromAbove #OutNetanyahu #OutPutin #OutMBS #OutTrump #PikuachNefesh8200 Talmud (Sanhedrin 37a); Matthew 16:26; Surah Al-Ma'idah (Chapter 5), Verse 32. #Future #Backwriters #BackwardsCausation #UnitedWeStand #TeamHumanity #ProHuman #Beloved ❤️🫶 #YouMatter #MeirDagan ܠܵܐ ܫܲ��ܡܵܐ ܐ݈ܢܵܐ ܠܲܝܟ݂، ܘܠܵܐ ܠܒܪܝܼܬ݂ܵܟ݂. ܒܲܪܬ݂ ܚܲܘܵܐ ܫܲܪܝܼܪܬܵܐ ܐ݈ܢܵܐ. ܠܵܐ ܝܕܲܥܬܝܼ ܠܐܸܡܝܼ؛ ܐܲܦܸܢ ܛܪܲܕܬܝܼ ܠܛܸܠܵܢܝܼܵܗ̇. ܡ̣ܢ ܢܘܼܗܪܵܐ ܕܩܵܕ݂ܡ ܒܪܝܼܬ݂ܵܐ ܐܬܹܝܬ݂.

Sara

@SMCADMAN

1 day ago

😳 @MarkJCarney @rcmpgrcpolice @CanadianForces @CanadianArmy @CJOC_COIC #LAPD #Palestine #Israel #Palantir #BlackRock #algorithms #LLM #WMDs #disgorgement Read: Technical methods of execution used by researchers and engineers and how to approach algorithmic disgorgement through various operational strategies outlined in ¹PNAS AI Model Disgorgement Study. 1. Machine Unlearning (Reactive): Using mathematical techniques and noise adjustment (influence functions) to make a trained model "forget" specific data points, though it balances heavy computing costs against model accuracy. 2. Data Compartmentalization (Proactive): Splitting training sets into isolated shards so that specific sub-models can be dropped or detached if a compliance issue arises. 3. Differential Privacy (Preemptive): Injecting mathematical noise during the initial training phase to limit how much any single data point influences the final model. ¹A. Achille, M. Kearns, C. Klingenberg, and S. Soatto, “AI model disgorgement: Methods and choices | PNAS,” PNAS, https://t.co/lA0WVzadHB (accessed Nov. 13, 2026). https://t.co/SejdG2XxKC Edited by Jeffrey Ullman, Stanford University (Retired), Stanford, CA; received July 12, 2023; accepted March 15, 2024.

0

1

0

177

0

25