TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Measuring the Progress of AI Research

154 pointsby blacksmythealmost 8 years ago

9 comments

Animatsalmost 8 years ago
I&#x27;d add robot manipulation in unstructured situations. Progress has been very slow, but has picked up a little in recent years. Stanford vision-guided robot assembly, 1973.[1] DARPA robot manipulation project, 2012.[2]<p>[1] <a href="https:&#x2F;&#x2F;archive.org&#x2F;details&#x2F;sailfilm_pump" rel="nofollow">https:&#x2F;&#x2F;archive.org&#x2F;details&#x2F;sailfilm_pump</a> [2] <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=jeABMoYJGEU" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=jeABMoYJGEU</a>
joe_the_useralmost 8 years ago
This seems like a compendium of metrics for processes which AI is a making progress on now. Doing something like that seems like a fine idea - I can&#x27;t judge the quality of these metrics but it&#x27;s hard to be excited by this.<p>However, what I think would be interesting would be for researchers to make a compendium of &quot;human abilities&quot;, classifying and quantifying them as well as possible. One could then analyze the progress which AI could make towards emulating those capacities.<p>Obviously, this would be a rather crude measure but it at least could give some idea of AI&#x27;s toward new capacities.
jamessbalmost 8 years ago
I made an alternative interface that displays the same data using D3: <a href="https:&#x2F;&#x2F;jamesscottbrown.github.io&#x2F;ai-progress-vis&#x2F;index.html" rel="nofollow">https:&#x2F;&#x2F;jamesscottbrown.github.io&#x2F;ai-progress-vis&#x2F;index.html</a>
shpxalmost 8 years ago
<a href="https:&#x2F;&#x2F;rodrigob.github.io&#x2F;are_we_there_yet&#x2F;build&#x2F;" rel="nofollow">https:&#x2F;&#x2F;rodrigob.github.io&#x2F;are_we_there_yet&#x2F;build&#x2F;</a> is something similar, although last update was in Feb 2016.
评论 #14686723 未加载
评论 #14686072 未加载
IshKebabalmost 8 years ago
There&#x27;s no way we have reached human-level performance on image recognition tasks. My guess is that when there is ambiguity (e.g. &#x27;ship&#x27; vs &#x27;boat&#x27;) the AI is better at learning the answer the labellers chose. Humans haven&#x27;t looked through the training data so they use their real-life biases which may not match those of the labellers.<p>Just a guess, but whether that is true or not we&#x27;re definitely not at human-level performance.
评论 #14687354 未加载
评论 #14691613 未加载
评论 #14690709 未加载
0x4f3759dfalmost 8 years ago
Harvard Professor, &quot;We are Building Artificial Brains and Uploading Minds to Cloud right now&quot; <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=amwyBmWxESA" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=amwyBmWxESA</a>
randcrawalmost 8 years ago
1) No robotics? No interaction with the physical world at all?<p>2) No measure of the AI&#x27;s ability to teach others? How can you say AI really understands if it can&#x27;t then teach 1) what it has learned, and 2) understand what essential facts a tyro does not know or misunderstands?<p>3) No assessment of the AI&#x27;s semantic interpretive skills, like those long emphasized by cognitive scientists, such as those in Doug Hofstadter&#x27;s &quot;Fluid Concepts and Creative Analogies&quot; -- i.e. Miller analogies, double entendres, literary symbolism, poetry interpretation, and so on?<p>Without mastery of analogies, an AI will have all the cultural insightfulness of a portrait of leisuresuit Elvis in neon paint on velvet.
评论 #14686731 未加载
ragebolalmost 8 years ago
I&#x27;d love to see a game with high complexity, long term planning, incomplete information etc in there as well, e.g. Star Craft.
评论 #14686766 未加载
narvindalmost 8 years ago
Great compilation!