Inferencing Big Data with AI and Machine Learning in Metaverse
summary
Claims realtime inferencing of metaverse data using standard ML techniques plus specialized infrastructure, with experiments that remove duplicates to find 18 customers identified and run k‑means for K=2–20 with 100 restarts to pick an optimal K. Uses R (Rattle), Hadoop/NoSQL/Spark backends, GPUs/TPUs and IBM’s Linux one for faster inferencing, and mentions Cascade/kisket quantum interfaces. Stakes include 180 zettabytes by 2025 and $10.5 trillion annual cybercrime projection, motivating hardware‑level security and quantum‑safe encryption.
AI-written summary · about
chapters (13)
- 0:00Intro
- 1:52About Gautam
- 3:39Why visualize big data
- 6:44How to visualize big data
- 7:29AR vs VR
- 8:38Virtual Fixtures
- 10:28Data Sources
- 13:11Machine Learning Concepts
- 15:28Removing Duplicate
- 16:48Predictive Functions
- 22:51Code Sniper 3
- 24:16Summary
- 26:15Question
related talks
15mMapping the Future of Data & AI
58mAI-Enhanced Big Data: Integrating Private LLMs and Vector Databases
30mAI: unleashing the next wave
29mUnlocking the Future of AI: Trends, Infrastructure, and Tokenomics
24mMachine Intelligence gets real
41mBig Data Serving: The Last Frontier. Processing and Inference at Scale in Real Time
23mFrom Gaming to the Metaverse
19mHow Intuit Streamlined AI/ML Inference Workflows on K8s