# Differentiable Reachability for Scalable Verification and Safe Training of Neural Network Controllers

**Budget Signal Score:** 55/100

| Field | Value |
|---|---|
| Institution | Georgia Tech Research Corporation |
| Location | ATLANTA, GA |
| Project | Differentiable Reachability for Scalable Verification and Safe Training of Neural Network Controllers |
| Award amount | $495,000 |
| Obligated so far | $495,000 |
| Awarded | 2026-08-06 |
| Period | 2027-01-01 to 2029-12-31 |
| Program | EPCL: Energy, Power, Control, |
| Directorate / division | ENG / ECCS |
| Award ID | 2625831 |

## Why this score
- Award of $300K or more.
- Equipment- and compute-heavy field.

Source: U.S. National Science Foundation Awards API (https://www.nsf.gov/awardsearch/showAward?AWD_ID=2625831). Last refreshed: 2026-11-03T18:10:42.804Z.
Award data comes from the NSF Awards API and may be amended by NSF. Institution-level only; investigator contact details are not included. Informational only.