
  <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
      <title>by aaron</title>
      <link>https://byaaron.tech/blog</link>
      <description>The open archive of my work</description>
      <language>en-us</language>
      <managingEditor>address@yoursite.com (Aaron)</managingEditor>
      <webMaster>address@yoursite.com (Aaron)</webMaster>
      <lastBuildDate>Sun, 01 Jun 2025 00:00:00 GMT</lastBuildDate>
      <atom:link href="https://byaaron.tech/feed.xml" rel="self" type="application/rss+xml"/>
      
  <item>
    <guid>https://byaaron.tech/blog/defi/swabbler</guid>
    <title>Using Machine Learning to Optimize Stablecoin Markets</title>
    <link>https://byaaron.tech/blog/defi/swabbler</link>
    <description>Example of a markdown file with code blocks and syntax highlighting</description>
    <pubDate>Sun, 01 Jun 2025 00:00:00 GMT</pubDate>
    <author>address@yoursite.com (Aaron)</author>
    <category>defi</category><category>algo</category><category>trading</category><category>arbitrage</category>
  </item>

  <item>
    <guid>https://byaaron.tech/blog/defi/hedging-impermanent-loss</guid>
    <title>Hedging against impermanent loss with Binance perpetual futures</title>
    <link>https://byaaron.tech/blog/defi/hedging-impermanent-loss</link>
    <description>A hedging strategy that uses long call and put positions on Binance perpetual futures to offset impermanent loss for a Uniswap V2 ETH/USDC liquidity provider. Includes the constant-product math behind the model and backtests showing reduced max drawdown in volatile and declining markets.</description>
    <pubDate>Mon, 13 Jan 2025 00:00:00 GMT</pubDate>
    <author>address@yoursite.com (Aaron)</author>
    <category>defi</category><category>hedging</category>
  </item>

  <item>
    <guid>https://byaaron.tech/blog/ml/titanic-gaussian-regression</guid>
    <title>Gaussian process regression on titanic dataset</title>
    <link>https://byaaron.tech/blog/ml/titanic-gaussian-regression</link>
    <description>Applying Gaussian process classification to the Titanic dataset to predict passenger survival from age and fare using scikit-learn and an RBF kernel. Walks through data exploration, preprocessing, and evaluation (~72% accuracy), with surface and heatmap visualizations of the predicted survival probabilities.</description>
    <pubDate>Fri, 13 Oct 2023 00:00:00 GMT</pubDate>
    <author>address@yoursite.com (Aaron)</author>
    <category>statistics</category><category>regression</category>
  </item>

    </channel>
  </rss>
