Strata Data Conference 2017 - New York, New York
2017-09-25..2017-09-28
New York, United States
[Strata Data Conference 2017 - New York, New York, Strata Data Conference 2018, Strata Data Conference 2018 - New York, NY, Strata Data Conference 2019 - London, United Kingdom, Strata Data Conference 2019 - New York, NY]
who it's for
data scientists, machine learning engineers, data engineers, platform engineers, cloud architects, engineering managers, product managers
topics
streaming personalization pipelines; aggregated mobile CDR mobility analysis; Xeon Scalable processor performance; distributed deep learning with BigDL; real-time model scoring in databases; persistent streaming platforms with JSON DBs; transfer learning and representation learning; model serving, drift detection and retraining; Mesos, HDFS and YARN runtime migration; risk terrain modeling for financial crime; IoT and smart city analytics; Apache Beam and Dataflow pipelines
| title | author | length |
|---|---|---|
| Music, the window into your soul | Christine Hung | 04m |
Spotify uses streaming counts and session signals plus machine learning to infer concrete life details and tune recommendations: models detect infers new… | ||
| Data science for the most vulnerable at UNICEF | Manuel García-Herranz | 02m |
UNICEF partners with a government and a telco to compute aggregated mobile CDRs that reveal movement flows during the 2014 Ebola crisis, showing that people… | ||
| Unleashing intelligence and data analytics at scale | Ma Ziya | 06m |
Intel presents end-to-end results: Hadoop and Spark on the new Xeon Scalable processors deliver 3x performance for big-data workloads, 2.7x for… | ||
| Teaching databases to learn in the world of AI | Nikita Shamgunov | 05m |
MemSQL presents a transactional database that consolidates tiers, runs transactions and analytics, and scores TensorFlow models in real time so applications… | ||
| A whole new way to think about your next-gen applications | Anil Gadre | 11m |
Argues that a converged platform (MapR) combines persistent streaming with billions of topics and a cloud-scale store so teams build real-time apps that run… | ||
| The age of machine learning | Ben Lorica | 04m |
Argues there are three bottlenecks: data, engineering, models and gives concrete fixes: prioritize labeled training sets, use transfer learning or unsupervised… | ||
| White Collar Crime Risk Zones | Sam Lavigne | 03m |
A predictive-policing style system uses parsed FINRA PDF enforcement records plus features like nonprofit, liquor store and investment-adviser densities to… | ||
| Journey to consolidation | Mike Olson, Cesar Delgado | 06m |
Siri migrates its runtime from heavy virtualization and filers onto Mesos and HDFS so a tiny operations team can run global real time ML at scale; binaries… | ||
| Wild, wild data | Jer Thorp | 03m |
Over six years a set of exploratory systems is built: EXO for spatial Kepler data, Convene for meeting networks across large organizations, and Specimen Box… | ||
| A tale of two cafeterias: Focus on the line of business | Tanvi Singh | 02m |
Compares Google outsourcing cafeterias to Stripe keeping food service in-house and claims that owning the cafeteria turns everyday transactions into a live… | ||
| Analytics everywhere, from things to cities | Raghunath Nambiar | 05m |
Predicts massive device growth and concrete city projects: claims 25 billion devices connected today, 50 billion by 2030 and urban growth to 662 cities with… | ||
| WTF? What's the future and why it's up to us | Tim O'Reilly | 03m |
Argues that technology should augment workers not automate them away, using Uber and Lyft as a concrete example: a matching platform plus Google Maps turns… | ||
| Human-AI interaction: Autonomous service robots | Manuela Veloso | 04m |
Manuela Veloso discusses autonomous service robots that learn from human sources and generate explanations in response to language-based requests about their… | ||
| The US EPA: Digital transformation through data science | Robin Thottungal | 02m |
A call to turn agency data into clear, evidence-first visuals: a century-long temperature plot that shows global average warming to about 1.5°C, and a NASA… | ||
| How the IoT and machine learning keep America truckin' | Terry Kline | 03m |
Claims that a connected fleet of 330,000+ connected vehicles is live and that integrating with 25 telematics providers powers remote diagnostics and predictive… | ||
| The real project of AI ethics | Joanna Bryson | 02m |
Uses artificially parsed word embeddings to show that semantic usage encodes social prejudice, reproducing human implicit bias patterns (faster associations of… | ||
| Your data is being manipulated. | danah boyd | 04m |
Argues that training data is both biased and actively targeted: shows ImageNet dataset artifacts and Latanya Sweeney’s finding that Google returned… | ||
| Implementing custom models for machine learning | Lucy Yu | 06m |
Announces MemSQL 6 with three concrete changes: enhanced online operations that keep mission‑critical data available, vectorized execution and compact query… | ||
| Building integrated data analytics ecosystem | Erin Banks | 07m |
Dell EMC consolidates Dell, VMware, Pivotal, RSA, SecureWorks and Virtustream into an integrated analytics stack and reports 140,000 employees coordinating… | ||
| Data science is trending towards open tooling and deep learning | Thomas Dinsmore | 08m |
Claims that enterprises shift from commercial packages to open tooling like R Python Scala and Spark and that interest in deep learning is exploding with… | ||
| How to not replicate on premise silos in the cloud | Nik Rouda | 09m |
Advice: avoid recreating on-premise silos by detaching persistent data and its security, audit and lineage from transient compute clusters, using cloud for… | ||
| Moving analytics from archive to active business intelligence | Jack Norris | 10m |
A new database announcement converges transactional and analytical workloads so one cluster handles high-read, high-write and in-place machine learning with… | ||
| The AI stack from chip to analytics | Ma Ziya | 09m |
Intel presents an end-to-end AI stack: client and server silicon (Xeon, Xeon Phi, FPGA) plus optimized software (MKL, inference SDKs) that runs on existing… | ||
| Will AI help save the snow leopard? | Joseph Sirosh | 12m |
A working pipeline trains a deep neural network (ResNet‑50 pretrained on ImageNet) to sort camera‑trap photos and deploys it on Azure as a Snow Leopard… | ||
| How the analytic economy is bringing value to all economies | Fiona McNeill | 09m |
Argues that an analytics economy extracts latent value from IoT, API and big‑data streams by turning data into improving decisions rather than just faster… | ||
| Elminating data silos with SDX | David Tishgart | 09m |
SDX (Shared Data Experience) promises a single governance layer that keeps policies travel with the data across S3, Azure Data Lake, HDFS and Kudu so ETL… | ||
| Democratizing the pipeline with Apache Beam and Dataflow | Tyler Akidau, Reuven Lax | 10m |
Describes Apache Beam as a portability framework and programming model for Java and Python pipelines and Google Cloud Dataflow as a managed runner that… | ||
| Building the best stack for the largest companies | Matt Winkler | 08m |
Microsoft describes how a platform layer ties tooling, services and apps so enterprises can move from experiments to production using concrete patterns: data… | ||
| Transforming AI into something real | Madhu Matta | 10m |
Lenovo presents an end-to-end AI stack from edge devices through PCs to data centers, combining high-performance and supercomputing hardware with open source… | ||
| UCS to help company transformation | Raghunath Nambiar | 14m |
UCS is Cisco’s unified computing system introduced 2009 and holds #1 market share for blade systems in North America and #2 worldwide. The claim is that… | ||
| The high performance stack | Prasad Venkatachar, Stewart Tate | 08m |
Claims a production integrated metal plus software stack built from Lenovo SR650 hardware, Intel processors, IBM software and Hortonworks that runs Hadoop and… | ||
| Brett Davis interviewed | Brett Davis | 02m |
Pharma companies use the recent explosion of electronic health care data plus AI/ML and deep learning as enablers to answer which treatments work for which… | ||
| Ryan Hoffmeister interviewed | Ryan Hoffmeister | 02m |
Argues that success comes from treating analytics as organizational change not just a new database, citing the EMR-driven data explosion and stricter privacy… | ||
| Raj Kamath interviewed | Raj Kamath | 05m |
Claims that practical adoption starts with data strategy and platform upfront and that most real stacks end up as a hybrid of commercial and open source, not… | ||
| AI for automation, algorithms, and intelligent agents | Ashish Verma | 10m |
Claims mainstream adoption of AI through self-driving cars using LiDAR and billion-dollar AV investments; reports working on autonomous stacks for OEMs and… | ||
| nothing matches | ||