index.xml

<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Trusted AI Ideas</title>
    <link>/</link>
    <description>Recent content on Trusted AI Ideas</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>en-us</language>
    <lastBuildDate>Wed, 14 Jun 2023 12:16:19 +0100</lastBuildDate>
    
	<atom:link href="/index.xml" rel="self" type="application/rss+xml" />
    
    
    <item>
      <title>EU AI Act explicitly mentions SMEs and start-ups, but how?</title>
      <link>/post/sme_startup_eu_ai_act/</link>
      <pubDate>Wed, 14 Jun 2023 12:16:19 +0100</pubDate>
      
      <guid>/post/sme_startup_eu_ai_act/</guid>
      <description>European Union is on the verge of adopting a horizontal Regulation on Artificial Intelligence (AI Act). Specific mentions of SMEs and start-ups are included. Let&amp;rsquo;s analyze them.
Here is the breakdown of the 30 explicit mentions, and their goals, of SMEs and start-ups in the AI Act Proposal:
 to smoothen some Provider and Deployer obligations (#12 mentions), to possibly reduce penalties when infringement occurs (#2), to facilitate access (process and price) to compliance processes (#7), to include them more as stakeholder in EU Governance (#4) and to affirm a will to support their innovation.</description>
    </item>
    
    <item>
      <title>Exploring ChatGPT guardails: from protected to forgotten countries (Part 1)</title>
      <link>/post/chatgpt_country_guardrails_study/</link>
      <pubDate>Fri, 24 Feb 2023 12:16:19 +0100</pubDate>
      
      <guid>/post/chatgpt_country_guardrails_study/</guid>
      <description>OpenAI ChatGPT has built guardrails on limiting the generation of negative content. How do these guardrails behave for countries? Spoiler: there are some gaps and disparities in ChatGPT safety mechanisms. Based on 24100 ChatGPT queries, this blog post is an exploration of ChatGPT responses when prompted to generate negative content about a country.
If you are in a hurry, go and see the early results.
1. Context: ChatGPT guardrails and harmful content prevention.</description>
    </item>
    
    <item>
      <title>Do you know the 4 types of additive Variable Importances?</title>
      <link>/post/variable_importance_feature_attribution/</link>
      <pubDate>Sat, 16 May 2020 15:36:11 +0200</pubDate>
      
      <guid>/post/variable_importance_feature_attribution/</guid>
      <description>Facing complex models, both computer simulation and machine learning practitioners have pursued similar objectives: to see how results could be broken down and linked to the inputs. Whether it is called Sensitivity Analysis or Variable Importance in the context of explainable AI, some of their methods share an important component: the Shapley values.
This article presents a structured 2 by 2 matrix to think about Variable Importances in terms of their goals.</description>
    </item>
    
    <item>
      <title>Breaking down factors of Covid-19 orientation algorithm by importance</title>
      <link>/post/covid_variable_importances_shapley/</link>
      <pubDate>Fri, 01 May 2020 15:15:56 +0200</pubDate>
      
      <guid>/post/covid_variable_importances_shapley/</guid>
      <description>Let&amp;rsquo;s dig deeper on the important factors of the French Covid-19 patient orientation algorithm! The short answer is fever, diarrhea and number of risk factors, but the real answer is that your target population matters a lot.
Follow me on this journey with 3 variable importance methods and get some insights about how factors interact.
If you are in a hurry, go and see the take-away messages or have a look at the python code.</description>
    </item>
    
    <item>
      <title>Simplifying the French Covid-19 patient orientation algorithm</title>
      <link>/post/covid_scoring_system_presentation/</link>
      <pubDate>Fri, 17 Apr 2020 08:51:53 +0200</pubDate>
      
      <guid>/post/covid_scoring_system_presentation/</guid>
      <description>Were you aware that there is an official French decision tree to orient patients towards relevant medical services, depending on their Covid-19 related symptoms? I transformed it into a point-based scoring system (100% logically equivalent), so that a patient can compute his/her score by adding simple weights. Using these Covid-19 orientation algorithms, a more general comparative analysis is carried out concerning decision trees and scoring systems.
If you are in a hurry, go and see the take-away messages!</description>
    </item>
    
  </channel>
</rss>