{"slug": "faster-fourier-transform", "title": "Faster Fourier Transform", "summary": "OpenAI posted a paper claiming an algorithm can compute the discrete Fourier transform in O(n (log n)^(1 − ε)) time for ε = 10^−13, a marginal asymptotic improvement over the Fast Fourier Transform's O(n log n). The Fast Fourier Transform remains the standard method for computing the discrete Fourier transform of a sequence of length n.", "body_md": "The Fast Fourier Transform (FFT) algorithm can compute the discrete Fourier transform of a sequence of length n in time O(n log n). OpenAI recently posted a paper saying there is an algorithm that could compute the discrete Fourier transform in O(n (log n)1 − ε) time for ε = 10−13. This result is amazing. It seemed that […]\n\nThe post", "url": "https://wpnews.pro/news/faster-fourier-transform", "canonical_source": "https://www.johndcook.com/blog/2026/10/07/faster-fourier-transform/", "published_at": "2026-10-07 21:41:47+00:00", "updated_at": "2026-10-07 22:17:42.911295+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-research"], "entities": ["OpenAI", "Fast Fourier Transform"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/faster-fourier-transform", "markdown": "https://wpnews.pro/news/faster-fourier-transform.md", "text": "https://wpnews.pro/news/faster-fourier-transform.txt", "jsonld": "https://wpnews.pro/news/faster-fourier-transform.jsonld"}}