<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: hadeson</title><link>https://news.ycombinator.com/user?id=hadeson</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 30 Apr 2026 02:49:15 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=hadeson" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by hadeson in "Poker Tournament for LLMs"]]></title><description><![CDATA[
<p>I don't think it's easier, a bad poker bot will lose a lot over a large enough sample size. But maybe it's easier to incorporate exploitation into your strategy - exploits that rely more on human psychology than pure statistics?</p>
]]></description><pubDate>Tue, 28 Oct 2025 11:11:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=45731359</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=45731359</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45731359</guid></item><item><title><![CDATA[New comment by hadeson in "Poker Tournament for LLMs"]]></title><description><![CDATA[
<p>From my experience, their hallucination when playing poker mostly comes from a wrong reading of their hand strength in the current state. E.g., thinking they have the nuts when they are actually on a nut draw. They would reason a lot better if you explicitly give out their hand strength in the prompt.</p>
]]></description><pubDate>Tue, 28 Oct 2025 08:47:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=45730496</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=45730496</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45730496</guid></item><item><title><![CDATA[New comment by hadeson in "Careless People"]]></title><description><![CDATA[
<p>"Magni, inquit, fures minorem puniunt" - Appeared in Pseudo-Caecilius Balbus.</p>
]]></description><pubDate>Fri, 25 Apr 2025 11:15:45 +0000</pubDate><link>https://news.ycombinator.com/item?id=43792383</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=43792383</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43792383</guid></item><item><title><![CDATA[New comment by hadeson in "g1: Using Llama-3.1 70B on Groq to create o1-like reasoning chains"]]></title><description><![CDATA[
<p>I like the Tree of Thoughts theory that treat each chain of thoughts 'branch' as a possible hypothesis. They might trained a search system that quickly explore some of these branches and by some metric choose the most likely to be the right one at the moment to answer.</p>
]]></description><pubDate>Mon, 16 Sep 2024 04:00:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=41552627</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=41552627</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41552627</guid></item><item><title><![CDATA[New comment by hadeson in "A $10k poker win changed how I think"]]></title><description><![CDATA[
<p>Up to six-player no-limit Hold’em poker with Pluribus.</p>
]]></description><pubDate>Tue, 07 Sep 2021 02:03:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=28440056</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=28440056</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=28440056</guid></item><item><title><![CDATA[New comment by hadeson in "Show HN: RentMyCPU – An open-source computational network"]]></title><description><![CDATA[
<p>This was a thing until mining ETH is profitable again.</p>
]]></description><pubDate>Wed, 14 Apr 2021 05:01:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=26803111</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=26803111</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=26803111</guid></item><item><title><![CDATA[New comment by hadeson in "Piipcam – 1080P IP Camera on a Raspberry Pi Zero W"]]></title><description><![CDATA[
<p>It seem this setup get close to the Jetson Nano 2Gb price range ($59), which from my experience is order of magnitude faster than rpi 4 for computer vision/video processing tasks. Is there any advantage of using Motion over nvidia-deepstream?</p>
]]></description><pubDate>Mon, 02 Nov 2020 03:18:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=24964543</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=24964543</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=24964543</guid></item><item><title><![CDATA[New comment by hadeson in "VkFFT – Vulkan Fast Fourier Transform Library"]]></title><description><![CDATA[
<p>It could be used to accelerate Convolutional Neural Nets training [0]<p>[0] <a href="https://arxiv.org/abs/1312.5851" rel="nofollow">https://arxiv.org/abs/1312.5851</a></p>
]]></description><pubDate>Mon, 28 Sep 2020 08:40:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=24614406</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=24614406</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=24614406</guid></item><item><title><![CDATA[New comment by hadeson in "Show HN: Build a DIY license plate reader with machine learning"]]></title><description><![CDATA[
<p>I guess object detection is the bottleneck of the pipeline, which highly optimized YOLOv4-tiny[0]could get to 39 FPS on Jetson Nano. Also camera decoding with Nvidia Deepstream is a lot faster than OpenCV.<p>[0] <a href="https://github.com/pjreddie/darknet/issues/2201" rel="nofollow">https://github.com/pjreddie/darknet/issues/2201</a></p>
]]></description><pubDate>Thu, 17 Sep 2020 03:41:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=24500966</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=24500966</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=24500966</guid></item><item><title><![CDATA[New comment by hadeson in "A Database of Historical Cookbooks"]]></title><description><![CDATA[
<p>Cooking is really an art with lots of tacit knowledge that vary too much to put in a simple recipe.</p>
]]></description><pubDate>Fri, 14 Aug 2020 09:38:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=24155776</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=24155776</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=24155776</guid></item><item><title><![CDATA[New comment by hadeson in "Show HN: Unscreen – Automatically remove video backgrounds with ML"]]></title><description><![CDATA[
<p>Most kinds of professional applications I could think of would better off with a green screen. On the other hand, using this for editing videos in the wild with messy background still not good enough.</p>
]]></description><pubDate>Wed, 05 Aug 2020 09:05:52 +0000</pubDate><link>https://news.ycombinator.com/item?id=24058696</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=24058696</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=24058696</guid></item><item><title><![CDATA[New comment by hadeson in "Fawkes: Image “Cloaking” for Personal Privacy"]]></title><description><![CDATA[
<p>Yandex seem to give even more precise matches than Google. Maybe image of someone less popular would work better.</p>
]]></description><pubDate>Thu, 23 Jul 2020 02:59:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=23923777</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=23923777</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=23923777</guid></item><item><title><![CDATA[New comment by hadeson in "Hackers take over prominent Twitter accounts in simultaneous attack"]]></title><description><![CDATA[
<p>Twitter investigation suggest that this is a coordinated social engineering attack [0]. The idea that the hackers are some non state actors and not from the US seem unlikely.
[0] <a href="https://twitter.com/TwitterSupport/status/1283591846464233474" rel="nofollow">https://twitter.com/TwitterSupport/status/128359184646423347...</a></p>
]]></description><pubDate>Thu, 16 Jul 2020 03:58:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=23855635</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=23855635</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=23855635</guid></item><item><title><![CDATA[New comment by hadeson in "What I learned from looking at every AI/ML tool I could find"]]></title><description><![CDATA[
<p>Really just curious about this approach and want to test it since most old scraping methods failed on Facebook data. My take is that it is possible with enough resources since it is actually pretty hard to separate this from real usages.</p>
]]></description><pubDate>Thu, 25 Jun 2020 04:32:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=23636937</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=23636937</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=23636937</guid></item><item><title><![CDATA[New comment by hadeson in "What I learned from looking at every AI/ML tool I could find"]]></title><description><![CDATA[
<p>I tried using OCR to scrap Facebook profiles by simulate web browsing behavior. It helps a lot in avoid account blocking but still too slow to be practical.</p>
]]></description><pubDate>Wed, 24 Jun 2020 07:20:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=23625048</link><dc:creator>hadeson</dc:creator><comments>https://news.ycombinator.com/item?id=23625048</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=23625048</guid></item></channel></rss>