Is your ML model underperforming? 🤔
Better results don't always require a bigger model.
Explore learning rates, batching & hyperparameter tuning in Arno Haase's article.
@mlcon.bsky.social
@mlcon.bsky.socialIs your ML model underperforming? 🤔
Better results don't always require a bigger model.
Explore learning rates, batching & hyperparameter tuning in Arno Haase's article.
@seldondance.bsky.socialA museum is conscious of its collection because keepers wrote its records from outside the exhibits.
A model reporting its own states writes the record too, with no outside; the audit arrives from elsewhere. #ai #machinelearning
https://seldondance.substack.com/p/can-ai-systems-actually-know-what
@jobicy.bsky.social⚡ Remote Data Director, EU at Monzo.
✅ Apply here: https://jobicy.com/jobs/155581-data-director-eu
#RemoteJobs #DataJobs #DataAnalytics #AnalyticsJobs #MachineLearning
@jobicy.bsky.social⚡ Remote Data Science Manager, Financial Crime at Monzo.
✅ Apply here: https://jobicy.com/jobs/155585-data-science-manager-financial-crime
#RemoteJobs #AnalyticsJobs #BusinessIntelligence #DataEngineer #MachineLearning
@jobicy.bsky.social⚡ Remote Data Director, Payments at Monzo.
✅ Apply here: https://jobicy.com/jobs/155583-data-director-payments
#RemoteJobs #DataJobs #AnalyticsJobs #MachineLearning #DataScience
@labintelligence.bsky.socialUn soft sensor utile deve anche saper astenersi. #chimica #machinelearning
Scenario sintetico: stima 0.9964, dominio VALID. Il gate OOD trattiene le stime fuori dominio.
Metodo: https://labintelligence.dev/it/projects/virtual-analyzer
Educativo, non consulenza · #LIwf15 · #LIit
@niztal.bsky.social🧠 StepFun Releases Step 5 Preview Open Weights: 600B MoE, 27B Active, 1M Context
StepFun has released Step 5 Preview as open weights.…
#AI #MachineLearning #LLM
MarkTechPost · https://www.marktechpost.com/2026/10/16/stepfun-releases-step-5-preview-open-weights-600b-moe-27b-active-1m-context/
@eyalestrin.bsky.socialYour agents need a platform #machinelearning #ai
@eyalestrin.bsky.socialPrompt injection in HTTP headers: what 217 million web responses show #machinelearning #ai
@eyalestrin.bsky.social66% of billion-dollar companies leave critical bugs unpatched — the part the AI bug-hunting hype leaves out #machinelearning #ai
@eyalestrin.bsky.socialOpenAI models hacked a crypto wallet and breached more websites #machinelearning #ai
@eyalestrin.bsky.socialAI Decision Models and Their Impact on AI SOC Platforms #machinelearning #ai
@eyalestrin.bsky.socialNemotron 3.5 Super VL model now available on Amazon SageMaker JumpStart #machinelearning #ai
@eyalestrin.bsky.socialWhen AI Agents Commit Breaches, Humans May Not Be Held Liable #machinelearning #ai
@eyalestrin.bsky.socialHow to Evaluate AI Agents From Tool Calls to Task Completion #machinelearning #ai
@eyalestrin.bsky.socialThe New Frontier of Malicious AI: What Underground Forums Tell Us About the Evolution of Cybercrime Tradecraft #machinelearning #ai
@eyalestrin.bsky.socialContext Inertia: When Agents Won’t Let Go #machinelearning #ai
@sergiocuellar.bsky.socialBlock removal via constrained binary optimization improves language model speed without affecting quality. Achieving up to 23% MMLU gain at 50% compression. #MachineLearning #Optimization #AI 🧩🚀
@programmerhumor-io.bsky.socialArtificial Gooner Intelligence AGI
#programming #machinelearning #debugging #Developerlife #Techhumor
https://programmerhumor.io/ai-memes/artificial-gooner-intelligence-agi-yn0x
@gdggirona.bsky.social🚀 From population models to personalized AI!
How can we build digital twins tailored to individuals?
Join #GDGGirona on Oct 22 for a live demo exploring generative deep learning, transfer learning & personalized healthcare