MediaWiki API result

{
    "compare": {
        "fromid": 1,
        "fromrevid": 1,
        "fromns": 0,
        "fromtitle": "Welcome to the Robustly Beneficial Wiki",
        "toid": 1,
        "torevid": 2,
        "tons": 0,
        "totitle": "Welcome to the Robustly Beneficial Wiki",
        "*": "<tr><td colspan=\"2\" class=\"diff-lineno\" id=\"mw-diff-left-l1\" >Line 1:</td>\n<td colspan=\"2\" class=\"diff-lineno\">Line 1:</td></tr>\n<tr><td class='diff-marker'>\u2212</td><td class='diff-deletedline'><div><del class=\"diffchange diffchange-inline\">&lt;strong&gt;MediaWiki has been installed.&lt;/strong&gt;</del></div></td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange diffchange-inline\">Welcome to the Robustly Beneficial wiki!!</ins></div></td></tr>\n<tr><td class='diff-marker'>\u00a0</td><td class='diff-context'></td><td class='diff-marker'>\u00a0</td><td class='diff-context'></td></tr>\n<tr><td class='diff-marker'>\u2212</td><td class='diff-deletedline'><div><del class=\"diffchange diffchange-inline\">Consult the [https://www.mediawiki.org/</del>wiki<del class=\"diffchange diffchange-inline\">/Special:MyLanguage/Help:Contents User's Guide] for information on using the wiki software.</del></div></td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange diffchange-inline\">== What is this </ins>wiki <ins class=\"diffchange diffchange-inline\">about? And why? ==</ins></div></td></tr>\n<tr><td class='diff-marker'>\u00a0</td><td class='diff-context'></td><td class='diff-marker'>\u00a0</td><td class='diff-context'></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">This wiki aims to list references, ideas and research questions in AI ethics. We hope to help better visualize the scope and the limits of current AI ethics.</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">The wiki is managed by the Lausanne Alignment Club, based in EPFL, Switzerland. Please check also our Robustly Beneficial YouTube channel, our Robustly Beneficial Podcast and our Robustly Beneficial Talks.</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">The wiki has just been launched, so most pages are still being written. But they will never be finished \u2014 this is the whole point of a wiki!</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">== New here? How about starting with these pages? ==</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">The pages can be roughly divided into 4 categories.</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">=== Why AI ethics is becoming critical ===</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">If you are new to AI ethics, you should probably start with the [[AI risks]] page. You could then go into arguably today's most important case of AI ethics, namely [[YouTube]]. Note that algorithms also offer formidable [[AI opportunities]] that are definitely worth considering. Find out more by reading about [[online polarization]], [[misinformation]], [[addiction]], [[mental health]] or [[hate]].</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">And if you know little about the current state of algorithmic research, you might want to check the latest [[impressive advances in AI]]. Or you could check some [[funny applications of AI]]. You can also read my [[rant against semantic debates]] and my [https://www.lesswrong.com/posts/bwqDrSZvhEDKxRf6z/a-rant-against-robots rant against robots].</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">=== How today's (and probably tomorrow's) AIs work ===</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">The most important principle of today's AI is surely [[machine learning]]. Today, it mostly relies on [[stochastic gradient descent]] for (deep) [[neural networks]], which allow [[representational learning]] (see [[convolutional neural network]], [[residual network]], [[LSTM]]). See also [[Turing 1950]], [[convexity]], [[generative adversarial network]], [[transformer]] and [[linear systems]].</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">[[Bayesianism]] has been argued to be the ideal form of supervised and unsupervised learning, if we had infinite computational power (see [[Solomonoff's demon]]). It has numerous desirable properties, like [[statistical admissibility]], [[Bayesian agreement]] or the [[Bayesian brain]] hypothesis. See also [[Bayesian examination]] and [[conjugate priors]].</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">A branch of learning called [[reinforcement learning]], which relies on [[Q-learning]] or [[policy learning]], seems likely to become the core framework of today's and tomorrow's AIs. [[AIXI]] achieves the upper-bound for [[Legg-Hutter intelligence]], which aims to measure [[artificial general intelligence|general intelligence]].</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">To understand the gap between Bayesianism/AIXI and practical machine learning, we need to understand the constraints of computational [[complexity]] theory. By building upon the [[Church-Turing thesis]] and knowledge from [[human brain computations]], this allows some insights into [[human-level AI]], in addition to [[experts' AI predictions]]. See also [[entropy]] and [[sophistication]].</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">AIs are already doing [[distributed learning]], which raises numerous challenges, like [[Byzantine fault tolerance]] and [[model drift]].</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">=== Why AI safety and ethics is harder than meets the eye ===</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">We want to get algorithms to do what we would really want them to do. But this turns out to raise numerous highly nontrivial problems, like [[Goodhart's law]], [[overfitting]], [[robust statistics]], [[confounding variables]], [[adversarial attacks]], [[algorithmic bias]], [[cognitive bias]], [[backfire effect]], [[distributional shift]], [[privacy]], [[interpretability]], [[reward hacking]], [[wireheading]] and [[instrumental convergence]]. Because of all such problems, it seems crucial that AIs be able to reason about their ignorance, using [[Bayesianism|Bayesian]] principles, [[moral uncertainty]] and [[second opinion querying]].</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">AI ethics also demands that we solve thorny philosophical dilemmas, like the [[repugnant conclusion]], [[Newcomb's paradox]] and [[moral realism]]. Unfortunately, we have numerous [[cognitive bias|cognitive biases]], which seem critical to understand to solve AI ethics. Results about [[counterfactual]], [[von Neumann-Morgenstern theorem]] and [[Dutch book]] also seem useful to consider.</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">=== How to solve AI ethics (hopefully) ===</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">To solve AI ethics, [http://ceur-ws.org/Vol-2301/paper_1.pdf Hoang][https://dblp.org/rec/bibtex/conf/aaai/Hoang19 19a] proposed the [[ABCDE roadmap]], which decomposes the [[alignment]] problem into numerous (hopefully) orthogonal and complementary subproblems. Such subproblems include [[data certification]], perhaps through [[Blockchain]], [[world model inference]] through [[Bayesianism]] and/or [[representational learning]], [[volition]] learning and [[social choice]] solutions, [[corrigibility]] and safe [[reinforcement learning]].</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">If you are interested in doing AI ethics research, please check out [[AI ethics research questions]].</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">== About the authors ==</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">L\u00ea holds a [https://publications.polymtl.ca/1668/ PhD in applied mathematics] from the \u00c9cole Polytechnique of Montreal. He did postdoctoral research at MIT, before joining EPFL as a science communicator. He runs EPFL YouTube channels [https://www.youtube.com/user/TheWandida Wandida] and [https://www.youtube.com/channel/UCfY6ovyFMaw30NRs-KrxrWw/ ZettaBytes], and French-speaking YouTube channels and podcasts [https://www.youtube.com/channel/UC0NCbj8CxzeCGIF6sODJ-7A Science4All], [https://www.youtube.com/channel/UCNHFiyWgsnaSOsMtSoV_Q1A/ Axiome] and [https://www.youtube.com/playlist?list=PLtzmb84AoqRQ0ikLb4yC4lKgjeDEIpE1i Probablement]. In 2020, he (will have) published &lt;em&gt;The Equation of Knowledge: From Bayes Rule to a Unified Philosophy of Science&lt;/em&gt; at CRC Press (French version is out [https://www.amazon.fr/Formule-du-Savoir-Hoang-Nguyen/dp/2759822605/ Hoang18&lt;sup&gt;FR&lt;/sup&gt;]). L\u00ea is still partly involved in research project. But his current main focus is synthesizing and organizing the current effort in AI ethics. </ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">In 2020, L\u00ea and Mahdi co-wrote the book &lt;em&gt;The Fabulous Endeavor: Make Artificial Intelligence Robustly Beneficial&lt;/em&gt; [https://laboutique.edpsciences.fr/produit/1107/9782759824304/Le%20fabuleux%20chantier HoangElmhamdi][https://scholar.google.ch/scholar?hl=en&amp;as_sdt=0%2C5&amp;q=Le+fabuleux+chantier%3A+Rendre+l%27intelligence+artificielle+robustement+b%C3%A9n%C3%A9fique&amp;btnG= 19&lt;sup&gt;FR&lt;/sup&gt;] (the English version is pending).</ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\"></ins></div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">&lt;!--</ins></div></td></tr>\n<tr><td class='diff-marker'>\u00a0</td><td class='diff-context'><div>== Getting started ==</div></td><td class='diff-marker'>\u00a0</td><td class='diff-context'><div>== Getting started ==</div></td></tr>\n<tr><td class='diff-marker'>\u00a0</td><td class='diff-context'><div>* [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list]</div></td><td class='diff-marker'>\u00a0</td><td class='diff-context'><div>* [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Configuration_settings Configuration settings list]</div></td></tr>\n<tr><td colspan=\"2\" class=\"diff-lineno\" id=\"mw-diff-left-l9\" >Line 9:</td>\n<td colspan=\"2\" class=\"diff-lineno\">Line 57:</td></tr>\n<tr><td class='diff-marker'>\u00a0</td><td class='diff-context'><div>* [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language]</div></td><td class='diff-marker'>\u00a0</td><td class='diff-context'><div>* [https://www.mediawiki.org/wiki/Special:MyLanguage/Localisation#Translation_resources Localise MediaWiki for your language]</div></td></tr>\n<tr><td class='diff-marker'>\u00a0</td><td class='diff-context'><div>* [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki]</div></td><td class='diff-marker'>\u00a0</td><td class='diff-context'><div>* [https://www.mediawiki.org/wiki/Special:MyLanguage/Manual:Combating_spam Learn how to combat spam on your wiki]</div></td></tr>\n<tr><td colspan=\"2\">\u00a0</td><td class='diff-marker'>+</td><td class='diff-addedline'><div><ins class=\"diffchange\">--&gt;</ins></div></td></tr>\n"
    }
}