We're thinking about taking a run at using Proximal Policy Optimization (PPO) to incorporate data accessed via retrieval augmented generation (RAG) into our LLM over time. Anyone else messing with this stuff?
Kraken Networks, Inc.’s Post
More Relevant Posts
-
Business Leader | Data Scientist | IoT, Blockchain & AI Strategist | Regulatory Affairs & Infrastructure Expert
I'm pleased to share with you the full version of the paper that I presented today at the Normalize Workshop on the Normative Design and Evaluation of Recommender Systems. Any comments / discussions that you would like to send me will be taken into account and acknowledged in the next paper I have committed to publish on the same subject. https://lnkd.in/d_5e6wYD
To view or add a comment, sign in
-
We added a lot of nice stuff. Try out subset row inequalities for set partitioning formulations (thanks to founders Bjørn Petersen and Simon Spoorendonk together with friends Mads Jepsen and David Pisinger), choose between a parallel pull or push algorithm for solving the resource constrained shortest path problem, enjoy dynamically adjusted neighbourhoods for NG-routes, and new pricing heuristics. Remember to follow us and try out the solver! 🤞 #orms #columngeneration #operationsresearch
To view or add a comment, sign in
-
enabling digital services for Student Loan related activities while maintaining the highest security standard, the most compliant personal data protection and customer-centric data-driven innovation.
I am pleased to share a new blog post titled "Road Surface Defect Detection -- From Image-based to Non-image-based: A Survey". This post delves into the critical issue of road surface defects and the evolving methodologies for their detection. The survey covers various road surface defect detection methods, categorizing them based on input data types and methodologies used. It also reviews non-image-based methods and discusses associated challenges and open problems. Read the full article at https://bit.ly/3uszuYl for valuable insights into this important field.
To view or add a comment, sign in
-
I just published Advanced RAG Retrieval Strategies: Flow and Modular
Advanced RAG Retrieval Strategies: Flow and Modular
ai.gopubby.com
To view or add a comment, sign in
-
Precise and fast semantic tagging of large document repositories is crucial for powerful knowledge discovery and RAG systems. Here is the solution to this challenge: https://lnkd.in/deNVE47K
To view or add a comment, sign in
-
-
New Webinar 🚨: Long-term memory with MemGPT Long-term memory for LLMs is an unsolved problem, and doing naive retrieval from a vector database doesn’t work. The recent iteration of MemGPT (Charles Packer et al.) takes a big step in this direction. Taking the LLM as an OS analogy (Andrej Karpathy), the authors propose “virtual context management” to manage both memory in-context window and in external storage. Recent advances in function calling allow these agents to read and write from these data sources, and modify their own context. We’re excited to host Charles Packer, Sarah Wooders et al. in our webinar this Friday 9am PT. Come check it out: https://lu.ma/c39w9ofy Source paper: https://lnkd.in/gg3JQHur
To view or add a comment, sign in
-
-
New preprint: Analysis and Implementation of Nonlocal Governing Operators with Local Boundary Conditions on a General Interval https://lnkd.in/gs-bDyje
To view or add a comment, sign in
-
-
Have you upgraded to StataNow yet? Unlock new features like: • High-dimensional fixed effects • Meta-analysis for correlations • Inference robust to weak instruments • And more Don’t miss out—click to learn more and upgrade today! https://buff.ly/3vXscwI
To view or add a comment, sign in
-