Medium publication on AI, machine learning and data topics, with articles from many contributors on GANs, LLMs, SQL and vector databases.
HN links it for individual articles, such as one on generating synthetic sequential data with GANs and another on the open-source Falcon-40B model.
Good catch—it's actually a hybrid approach:
1. Anomaly Detection & Visualisation: This runs 100% locally via WASM using the linear-time algorithms I shared here previously.
2. Forecasting: This hits our DriftMind API [1][2].
We designed the API to be transient-only. The data is processed in-memory for the forecast and immediately discarded; absolutely nothing is persisted to disk or used for training. We wanted to keep the browser lightweight while still offering the heavy-duty forecasting from our core engine
[1] https://thingbook.io/doc/devGuide.html
[2] https://medium.com/towards-artificial-intelligence/reflexive-memory-a-cpu-only-alternative-to-transformers-for-streaming-forecasting-45efc12e383c
Very cool!
We’re using GANs to generate synthetic transactional data that preserves temporary and causal correlations [0].
[0] friends link to avoid paywall: https://medium.com/towards-artificial-intelligence/generating-synthetic-sequential-data-using-gans-a1d67a7752ac?source=friends_link&sk=1ab69626e19b95556cc3b2db83e64cf9
https://medium.com/towards-artificial-intelligence/falcon-40b-a-fully-opensourced-foundation-llm-945dd9824157