Overview
What STRATA Does
STRATA estimates the most likely race/ethnicity of individuals from name and US address for aggregate, population-level analysis such as HMDA/ECOA/CRA fair-lending monitoring, housing-equity research, and disparity studies. It is not licensed for decisions about identified or identifiable individuals - see the EULA for permitted-use and prohibited-use terms.
The model uses a published char-LSTM + geography architecture refined by gradient-boosted soft-voting, improving on traditional BISG/BIFSG approaches (see the held-out comparison under "How It Improves"). A geography waterfall selects the best-supported tier per record (block-group+tract, tract, ZCTA, or name-only) and reports the lookup level used.
Output Fields
Each scored record returns:
- Probabilities for five classes: White, Black, Hispanic, Asian, Other
- Argmax label (most likely class)
- Per-record uncertainty index (IU) - addresses a known BISG limitation where no per-record confidence is available. The uncertainty index was calibrated on Florida, Georgia and North Carolina voter records and PPP borrowers; its calibration in other populations is not established.
- Privacy-preserving cohort tokens (cohort_tract, cohort_zcta) for geographic grouping without exposing census identifiers
- Lookup level and quality codes indicating which geographic tier was resolved
Resolved geography is withheld from output to protect privacy.
How It Improves on BISG/BIFSG
- Character-level deep learning captures name-origin signals that surname-list approaches miss
- Geography waterfall uses the most granular available tier rather than a single fixed level
- Per-record uncertainty (IU) lets you flag low-confidence predictions for review - standard BISG provides no such signal
- Lower false-positive bias: On a held-out voter-registration sample, STRATA's block-group tier assigned 18.0% of non-white individuals to white, versus 28.5% for surname + ZIP-code BISG.
Privacy and Security
STRATA deploys as a SageMaker real-time endpoint or Batch Transform entirely within your AWS account. Your input PII never egresses to the vendor - the vendor never sees your data. The model inherits your existing AWS security posture including encryption in transit (TLS) and at rest (S3 server-side encryption for model artifacts). You control VPC configuration, IAM permissions, and network isolation.
Deployment Notes
- Model package size: approximately 1.45 GB
- Startup time (version 1.0.1): the container loads all models and geographic tables at startup and reports ready only when they are loaded; set ContainerStartupHealthCheckTimeoutInSeconds to 900 for real-time endpoints. On ml.m5.xlarge an endpoint reaches InService about 9 minutes after creation (measured: 524 s), when loading is complete, and the first request is then answered normally (measured: 7 s).
- Batch Transform: recommended for one-off or batch scoring to avoid standing hourly endpoint costs.
Getting Started
After subscribing, deploy the model package to a SageMaker endpoint or configure a Batch Transform job. Input records require name and US address fields. Contact the aequumAI team at the support email to request a methodology walkthrough or discuss a pilot scoring engagement for your use case.
About aequumAI
aequumAI builds privacy-preserving demographic estimation tools for regulated industries. STRATA builds on the methodology described in arXiv:2504.21259. The deployed model is trained on a larger corpus than the paper's evaluation models. It is designed for fair-lending compliance workflows where accuracy and auditability are critical. Methodology reference: S. Chalavadi, A. Pastor, and T. Leitch, arXiv:2504.21259 (https://arxiv.org/abs/2504.21259 ).
Highlights
- Beyond BISG/BIFSG: STRATA's char-LSTM + geography waterfall captures name-origin signals that surname-list approaches miss. On a held-out voter-registration sample, its block-group tier assigned 18.0% of non-white individuals to white vs 28.5% for surname + ZIP-code BISG. Each record carries an uncertainty index (IU) that standard BISG lacks; IU was calibrated on FL, GA and NC voters and PPP borrowers, and its calibration elsewhere is not established.
- Privacy by architecture: Deploys entirely in your AWS account as a SageMaker real-time endpoint or Batch Transform. Your input PII never leaves your account and the vendor never sees your data. Resolved geography is withheld from output; privacy-preserving anonymous cohort tokens (cohort_tract, cohort_zcta) let you group results by geography without exposing census identifiers.
- Licensed for aggregate fair-lending and disparity analysis ONLY (HMDA/ECOA/CRA monitoring, housing-equity research) - not for individual decisions. Output includes 5-class probabilities (White, Black, Hispanic, Asian, Other), argmax label, uncertainty index, cohort tokens, and lookup-level quality codes. See EULA for permitted-use and prohibited-use terms.
Details
Features and programs
Financing for AWS Marketplace purchases
Pricing
Dimension | Description | Cost/host/hour |
|---|---|---|
ml.m5.xlarge Inference (Real-Time) Recommended | Model inference on the ml.m5.xlarge instance type, real-time mode | $13.58 |
ml.m5.xlarge Inference (Batch) Recommended | Model inference on the ml.m5.xlarge instance type, batch mode | $13.58 |
ml.m5.2xlarge Inference (Real-Time) | Model inference on the ml.m5.2xlarge instance type, real-time mode | $24.33 |
ml.m5.2xlarge Inference (Batch) | Model inference on the ml.m5.2xlarge instance type, batch mode | $24.33 |
Dimensions summary
Top-of-mind questions for buyers
Vendor refund policy
Software charges are metered per instance-hour through AWS Marketplace. Contact terry@aequum.ai within thirty (30) days of an invoice for a billing adjustment for a verified duplicate charge, metering error, or material inability to use the product caused by a confirmed defect; approved adjustments are issued through AWS Marketplace. AWS infrastructure charges are governed solely by AWS.
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
Amazon SageMaker model
An Amazon SageMaker model package is a pre-trained machine learning model ready to use without additional training. Use the model package to create a model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Version release notes
Startup fix for real-time endpoints. The container now loads all models and every state's geographic table before reporting ready (/ping returns 503 until loading is done), so the first request after InService no longer times out. Model weights, resolver tables and outputs are unchanged. The model archive now includes the Apache-2.0 NOTICE and LICENSE for the geographic lookup tables derived from Zest AI's ZRP, and no longer includes the unused crest1 package. Set ContainerStartupHealthCheckTimeoutInSeconds to 900. Batch Transform is recommended for one-off scoring.
Additional details
Inputs
- Summary
CSV with a header row; one individual per row. Columns: fname,mname,lname,address,city,state,zip5. fname and lname are required; mname optional. The name drives the character-LSTM; the US address adds geographic signal via an offline resolver. Any extra columns (e.g. a record id) pass through unchanged. US scope.
- Input MIME type
- text/csv
Support
Vendor support
Contact Support
For questions about deployment, model methodology, or permitted use, contact the aequumAI team at terry@aequum.ai .
Support Scope
Support covers assistance with SageMaker endpoint deployment, Batch Transform configuration, input/output format questions, model interpretation guidance, and licensing clarification.
For refund requests, contact the same email address with your AWS account ID and subscription details.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products
