2023年1月18日

美國東部標準時間2023年1月18日上午09:00舊金山——(美國商業資訊)——以數據為中心的AI平台公司Snorkel AI今天宣布了創紀錄的2022年業務勢頭，由三數字的客戶增長引領。新加入的客戶包括紐約梅隆銀行、黑天鵝數據、喬治敦大學安全和新興技術中心(CSET)和Pixability。Snorkel AI還將美國前十大銀行中的五家納入其客戶組合，並將其名單擴大到包括金融服務、保險、製藥和醫療保健、製造和零售行業的一些政府機構和財富500強公司。企業數據科學和機器學習團隊使用該公司的旗艦AI平台Snorkel Flow，該平台由程序化標記驅動，可將AI應用程序開發速度提高10-100倍。組織在每個項目上獲得了數十萬到超過百萬美元的投資回報，並為NLP、文檔處理、自動化、分析以及更多被手動標記阻止的用例解鎖了全新的用例。“人工智能的發展速度正在加快，特別是在基礎模型方麵，”Snorkel AI的首席執行官兼聯合創始人亞曆克斯·拉特納(Alex Ratner)說。“然而，比以往任何時候都多的企業被為其特定用例和生產質量要求調整模型所需的培訓數據有效地阻止了。憑借我們以數據為中心的AI平台，我們彌合了企業AI和最新創新之間的差距。我們的市場領導地位和增長清楚地表明，這為我們的客戶釋放了價值。”Snorkel AI在2022年繼續創新，宣布了Snorkel Flow的普遍可用性。 The company also added a number of data-centric AI capabilities designed to accelerate enterprise AI development with automated labeling, efficient collaboration between data scientists and subject matter experts, and rapid, model-guided iteration. Data-centric Foundation Model Development: Workflows to fine-tune foundation models or large language models and use them to build smaller, specialized models deployable for complex, performance-critical enterprise use cases. Continuous Model Feedback: Automated model training and integrated analysis to provide near real-time guidance when developing training data. Cluster View: Instant creation of embedding-based clusters and auto-generated cluster labeling functions to quickly turn a massive unlabeled dataset into high-quality training data. New Studio: Support for a wide range data types and ML tasks such as text and document classification and extraction, PDF and conversational AI pipelines, sequence tagging, and more. Additional 2022 highlights included: Expanded Partnership Ecosystem: Snorkel AI deepened partnerships across the enterprise AI ecosystem including Microsoft Azure AI Services, Google BigQuery and Vertex AI, Snowflake, Aimpoint Digital, Four Inc. and more. Hosted Largest Data-Centric AI Conference: Future of Data-Centric AI 2022 brought together thousands of leaders of data science and analytics teams, AI/ML practitioners, researchers, and students. The three-day event featured over 50 sessions from Apple, AstraZeneca, CIA, Capital One, Comcast, Databricks, Ford, Google, JPMorgan Chase, Meta, NVIDIA, Orange, State Farm, and more. Published Breakthrough Research: Snorkel AI Research published over 30 papers at NeurIPS, ICLR, ACL and more with academic partners at Brown University, Stanford University, University of Washington, and University of Wisconsin-Madison. The research team also contributed to major initiatives such as the WRENCH benchmark and BigScience advancing the state-of-the-art for data-centric AI. About Snorkel AI Founded by a team spun out of the Stanford AI Lab, Snorkel AI makes AI application development fast and practical by unlocking the power of machine learning without the bottleneck of manually-labeled training data. Snorkel Flow is the first data-centric AI platform powered by programmatic labeling. Backed by Addition, Greylock, GV, In-Q-Tel, Lightspeed Venture Partners and funds and accounts managed by BlackRock, the company is based in Palo Alto. For more information on Snorkel AI, please visit: https://www.snorkel.ai/ or follow @SnorkelAI. Contacts