Machine Learning Assisted Electronic Monitoring

The NOAA solicitation seeks a contractor to test, benchmark, and improve existing machine‑learning algorithms that automatically identify marine species and estimate their lengths from electronic monitoring data. The effort will involve evaluating current model performance, developing enhancements, and delivering validated software or model updates that can be integrated into NOAA’s monitoring systems. • Perform comprehensive testing of current species‑identification and length‑estimation algorithms. • Benchmark algorithm accuracy against established datasets and field observations. • Research and implement improvements (e.g., new model architectures, training data augmentation, hyper‑parameter tuning). • Deliver documentation, code, and validation reports demonstrating performance gains. • Provide knowledge transfer or training to NOAA staff as needed.