BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Huntsville AI - ECPv6.8.3//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Huntsville AI
X-ORIGINAL-URL:https://www.hsv.ai
X-WR-CALDESC:Events for Huntsville AI
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:20250309T080000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:20251102T070000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250205T180000
DTEND;TZID=America/Chicago:20250205T190000
DTSTAMP:20260424T205832
CREATED:20241216T012527Z
LAST-MODIFIED:20250203T052119Z
UID:1816-1738778400-1738782000@www.hsv.ai
SUMMARY:DeepSeek-R1 and Other Topics
DESCRIPTION:This week we will be back at HudsonAlpha in their classroom. This is a new room for us\, which I believe is on the opposite side from the large conference room we have been using. \nHudsonAlpha Tech Challenge \nWe should have Tyler Clark stop by for a few minutes to discuss some changes to HATCH and how we can get involved. Otherwise\, I can cover based on what I know. The main change from the past challenges is that this will be a one day challenge occurring on Saturday\, March 1st. \nDeepSeek-R1 \nWe’ll be discussing DeepSeek-R1\, which has been unavoidable in most conversations over the last week – and rightly so. We will cover the architecture\, the approach used for training\, as well as the distilled models based on the Llama and Qwen architectures. We have several folks in the group that already have DeepSeek-R1 running locally\, so if you need help getting it running or would like to connect directly please let me know. \nFor part of the discussion\, you may want to refer back to the Mixture of Experts talk that Josh Phillips led during one of our meetups in October 2023. You can find the link below. \nPaper Review Series \nThis year we are going to start a review series to look at published papers in-depth. These will be a monthly\, virtual-only meetup to start with but may shift based on interest. The first paper review will be on Wednesday\, February 12\, led by Josh Phillips to cover the DeepSeek-R1 paper (also linked below). \nSBIR/STTR Topics? \nThe next round of SBIR/STTR Topics are supposed to drop on Wednesday. Given the issues with funding being stopped for NSF grants and other changes driven by the new administration\, I don’t know if we will see new topics drop or not. \nAI Symposium \nWe’ll take a few minutes to talk about how the AI Symposium sessions went.  The organizers are interested in feedback for the quality of the sessions and any suggested changes for next year.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks & Other Events: \n\nDeepSeek-R1 on GitHub – https://github.com/deepseek-ai/DeepSeek-R1?tab=readme-ov-file\nDeepSeek-R1 Paper – https://arxiv.org/pdf/2501.12948\nMixture of Experts Discussion – https://hsv.ai/2023/10/25/mixture-of-experts-harnessing-the-hidden-architecture-of-gpt4/\nIntroduction to AI Slides – https://github.com/HSV-AI/presentations/blob/master/2025/250116_Introduction_to_AI_LearningQuest.pdf\nUses of AI Slides – https://github.com/HSV-AI/presentations/blob/master/2025/250123_Uses_of_AI_LearningQuest.pdf?raw=true\n2025 HATCH – https://hudsonalpha.org/techchallenge/\n2025 BSides – https://nac-issa.org/events/bsides/\n\nDetails: \n\nDate – 2/05/2025\nTime – 6-7:30pm\nLocation – HudsonAlpha\nAddress –  601 Genome Way Northwest\, Huntsville\, AL 35806\nZoom –https://us02web.zoom.us/j/82283427713?pwd=PCuFpoeIIYQGhAWDain4YK4FoRbETU.1
URL:https://www.hsv.ai/event/deepseek-r1/
LOCATION:HudsonAlpha\, 601 Genome Way Northwest\, Huntsville\, AL\, 35806
ATTACH;FMTTYPE=image/png:https://www.hsv.ai/wp-content/uploads/2024/12/DeepSeek-R1.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250212T180000
DTEND;TZID=America/Chicago:20250212T193000
DTSTAMP:20260424T205832
CREATED:20250209T145141Z
LAST-MODIFIED:20250209T145141Z
UID:1851-1739383200-1739388600@www.hsv.ai
SUMMARY:Virtual Paper Review - Reasoning
DESCRIPTION:Join us virtually this Wednesday at 6pm to kick off our monthly Paper Review series! We will be holding these sessions every 4 weeks on Zoom. The first one will be led by Josh Phillips: \nThis month we will be exploring the current evolution in AI reasoning models. We will start with a short review of the history of combining reinforcement learning with tree search for reasoning (AlphaGo/MuZero). Then we will cover the real start of the reasoning race with the chain-of-thought innovations highlighted by “Let’s Verify Step by Step.” And finally of course\, we’ll cover the most important insights from DeepSeek’s three recent papers that build on all of this prior foundation: \n\nDeepSeek-V3 – a highly efficient MoE-based approach (671B parameters\, only 37B active per token) powered by advanced CUDA/NCCL optimizations and low-precision (FP8) training\, making huge models tractable on constrained hardware.\nDeepSeek-R1 – the first open demonstration that an LLM can learn complex reasoning purely through RL rewards\, without any annotated examples (r1-zero)\nDeepSeekMath– which introduces GRPO (an alternative to PPO that simplifies training and removes the need for a separate value network).\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks: \n\nMastering Atari\, Go\, Chess and Shogi by Planning with a Learned Model (2019) :  https://arxiv.org/pdf/1911.08265\nLets Verify Step by Step (2023) : https://arxiv.org/abs/2305.20050\nDeepseek V3 Technical Report (2024): https://github.com/deepseek-ai/DeepSeek-V3/blob/main/DeepSeek_V3.pdf\nDeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning  (2025) https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf\nDeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models (2025): https://arxiv.org/pdf/2402.03300\n(Bonus) “We’ve got R1 at home” : https://unsloth.ai/blog/r1-reasoning\n\nDetails: \n\nDate – 2/12/2025\nTime – 6-7:30pm\nLocation – VIRTUAL\nZoom –https://us02web.zoom.us/j/86964741971?pwd=SXmv4xfY8M6TZbEOIXyYfLu5Y3FAil.1
URL:https://www.hsv.ai/event/virtual-paper-review-reasoning/
ATTACH;FMTTYPE=image/png:https://www.hsv.ai/wp-content/uploads/2025/02/Paper-Series-Reasoning.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20250219T180000
DTEND;TZID=America/Chicago:20250219T190000
DTSTAMP:20260424T205832
CREATED:20241216T012737Z
LAST-MODIFIED:20250219T211911Z
UID:1818-1739988000-1739991600@www.hsv.ai
SUMMARY:VIRTUAL ONLY - Deep Research Overview
DESCRIPTION:This week we will have Jack Francis provide an overview of Deep Research: \nWe will explore Google’s Gemini Deep Research\, OpenAI’s Deep Research\, and Open-Source Deep Research implementations. Deep Research is an LLM agent that uses reasoning to synthesize large amounts of online information and complete multi-step research tasks for you. We’ll dive into the mechanics of Deep Research\, discussing how these agents function and some of the benchmarks used to evaluate the models. \nWe will discuss the strengths and limitations of each approach\, provide examples for comparison\, and talk through Deep Research use cases\, such as market research! Deep Research is useful tool to add to your toolbox for market research\, decision analysis\, or any other task that requires significant amounts of “Googling” today. The goal of the presentation is to introduce Deep Research and provide an overview of current tools\, so that you can use it effectively for your own research tasks. \nPaper Review Series \nThanks to Josh for the great kickoff to the paper review series last week! You can find the recording\, links to papers\, and transcript here: \nhttps://hsv.ai/2025/02/12/ai-reasoning-reinforcement-learning-fundamental-papers-history/ \nIf you have a paper that you would like to see us review\, please let me know. We’ll post the date for March as soon as it’s finalized.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nLinks & Other Events: \n\n2025 HATCH – https://hudsonalpha.org/techchallenge/\n2025 BSides – https://nac-issa.org/events/bsides/\n\nDetails: \n\nDate – 2/19/2025\nTime – 6-7:30pm\nLocation – VIRTUAL\nZoom –https://us02web.zoom.us/j/84476860693?pwd=D2UGBsfpb5ZKoJEGMiFZaroV9aEb28.1
URL:https://www.hsv.ai/event/deep-research-overview/
ATTACH;FMTTYPE=image/png:https://www.hsv.ai/wp-content/uploads/2024/12/Deep-Research-Overview.png
END:VEVENT
END:VCALENDAR