Detection/Classification/Identification
Black River researches and applies the latest machine learning, artificial intelligence and data science techniques to our applied solutions. We have implemented time series models and algorithms such as Long Short-Term Memory (LSTM), detection models such as Convolutional Neural Networks (CNN), combination architectures such as CNN-LSTMs, Variational Auto-Encoders (VAEs), Generative Adversarial Networks (GANs), Reinforcement Learning (RL), Text Embedding Sentence Transformers, and Large Language Models (LLMs). These algorithms have been implemented across PyTorch and TensorFlow frameworks and deployed to CPU, GPU, and Multi-GPU based systems. Instantiations span small footprint edge devices to large high-side servers.