One Article Review
| Source |  ProofPoint |
|---|---|
| Identifiant | 8386760 |
| Date de publication | 2023-09-22 05:00:22 (vue: 2023-09-22 16:02:53) |
| Titre | Nébuleuse: une plate-forme ML de nouvelle génération Nebula: A Next-Gen ML Platform |
| Texte | Engineering Insights is an ongoing blog series that gives a behind-the-scenes look into the technical challenges, lessons and advances that help our customers protect people and defend data every day. Each post is a firsthand account by one of our engineers about the process that led up to a Proofpoint innovation. Cyber threats are increasing in their frequency and sophistication. And for a cybersecurity firm like Proofpoint, staying ahead of threats requires us to deploy new machine learning (ML) models at an unprecedented pace. The complexity and sheer volume of these models can be overwhelming. In previous blog posts, we discussed our approach to ML with Proofpoint Aegis, our threat protection platform. In this blog, we look at Nebula, our next-generation ML platform. It is designed to provide a robust solution for the rapid development and deployment of ML models. The challenges We live and breathe supervised machine learning at Proofpoint. And we face active adversaries who attempt to bypass our systems. As such, we have a few unique considerations for our ML process: Speed of disruption. Attackers move fast, and that demands that we be agile in our response. Manual tracking of attacker patterns alone isn\'t feasible; automation is essential. Growing complexity. Threats are becoming more multifaceted. As they do, the number of ML models we need escalates. A consistent and scalable modeling infrastructure is vital. Real-time requirements. It is essential to block threats before they can reach their intended targets. To be effective on that front, our platform must meet unique latency needs and support optimized deployment options for real-time inference. In other ML settings, like processing medical radiographs, data is more stable, so model quality can be expected to perform consistently over time. In the cybersecurity setting, we can\'t make such assumptions. We must move fast to update our models as new cyber attacks arise. Below is a high-level overview of our supervised learning process and the five steps involved. A supervised learning workflow, showing steps 1-5. Data scientists want to optimize this process so they can bootstrap new projects with ease. But other stakeholders have a vested interest, too. For example: Project managers need to understand project timelines for new systems or changes to existing projects. Security teams prefer system reuse to minimize the complexity of security reviews and decrease the attack surface. Finance teams want to understand the cost of bringing new ML systems online. Proofpoint needed an ML platform to address the needs of various stakeholders. So, we built Nebula. The Nebula solution We broke the ML lifecycle into three components-modeling, training and inference. And we developed modular infrastructure for each part. While these parts work together seamlessly, engineering teams can also use each one independently. The three modules of the Nebula platform-modeling, training and inference. These components are infrastructure as code. So, they can be deployed in multiple environments for testing, and every team or project can spin up an isolated environment to segment data. Nebula is opinionated. It\'s “opinionated” because “common use cases” and “the right thing” are subjective and hence require an opinion on what qualifies as such. It offers easy paths to deploy common use cases with the ability to create new variants as needed. The platform makes it easy to do the right thing-and hard to do the wrong thing. The ML lifecycle: experimentation, training and inference Let\'s walk through the ML lifecycle at a high level. Data scientists develop ML systems in the modeling environment. This environment isn\'t just a clean room; it\'s an instantiation of the full ML lifecycle- experimentation, training and inference. Once a data scientist has a model they like, they can initiate the training and inference logic in the training environment. That environment\'s strict polici |
| Notes | ★★★ |
| Envoyé | Oui |
| Condensat | ability able about accelerate access account across actionable active actors adam add address advances adventurous adversaries aegis after agile agility ahead ahead ai/ml alarms alec all allow allowing allows alone also always analysis apache application applications approach appropriate approval approved archetype architect are areas arise art artificial arts asset associate assumptions assurance attack attacker attackers attacks attempt authors automated automation autoscaling available avoid away aws bachelor back backend based because becomes becoming been before behind being below best big blend block blog blue bootstrap box branch branches breathe breed bring bringing brings broke build building built bundled business but bypass can canary candidate career case cases cases” catch challenges challenges change changes charge check christian ci/cd clean cloud code codebase collaborating collaboration collaborative college combined commits committed common commonalities company complete complex complexity component components computer computing conditions configurations considerations consistent consistently consolidation constantly continue contracts contributors control controls core correct corresponding cost costs coupled coupling covered covers create crime custom customer customers customizable: customization cyber cybersecurity dashboards data day decrease default defaults defend degree delegated delivering demand demands deploy deployed deploying deployment deployments design designed detailed detection develop developed developing development develops devops different discussed disruption distributed dive drift duke each ease ease easily easy ecosystem edge effective efficiency eligible email empowered empowering enable end endpoint endpoints enforced enforces engineer engineering engineers enhanced ensure entire environment environments escalates essential evaluated even every evolving exabytes example: existing expected experience experiment experimentation expert expertise exploratory extensible extracts face far: fast faster feasible; feature features features: field fighting finance firm firsthand five flexibility flexible flow fluid focus focused forefront fork freedom frequency from front full fully functionality gen generation git gives goal governance green group growing guarantees handle handles hands hard harvey has have help helps hence here high his hit holds ideal illustrates improve include: includes increasing incredible independently industry inference inference infrastructure initially initiate innersource innovation innovative insight insights instantiation integrated integration intelligence intended interconnected interest interested internal invaluable involved isn isolated issues its jobs join joun jump just keep key lab landscape latency launches lead leaders leadership leading leads learn learning lebedev led lessons let lets level lifecycle lifecycle: like lineage live logic look looking machine main maintainability make makes making manage managed managers manages manual manually many massive mastery math mathematical mathematics maven maximize may mba means meanwhile medical meet met metrics minimize minors minutes mlops model modeling modeling models modular modularity modules monitoring more move moving mudd multifaceted multiple must native nebula nebula: need needed needs new next not novel now number numerous observability offer offerings offers once one ongoing online open operational operations opinion opinionated opportunities optimize optimized optimizing options orchestration other out outcomes outcomes over overhead overview overwhelming ownership pace paradigm parameters part parts paths patterns people per perform phase pipeline pipelines place places platform platforms platform plus podcast policies pomona positive post posts practice practices predictable prefer prevention previous principles privacy problems process process: processes processing product production productivity products programmatically project projects promote proofpoint |
| Tags | Threat Medical Cloud |
| Stories | |
| Move | 
|
L'article ne semble pas avoir été repris aprés sa publication.
L'article ne semble pas avoir été repris sur un précédent.