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.

AmpliGraph: A TensorFlow-Based Library for Knowledge Graph Embeddings

112 pointsby mulletboyabout 6 years ago

6 comments

nlabout 6 years ago
Graph embeddings are one of my favorite underused things in ML.<p>I&#x27;ve used them to do things like characterise users based on follow&#x2F;follower patterns, but there are many more applications.<p>In the past I&#x27;ve had great success with Facebook Research&#x27;s StarSpace
评论 #19577883 未加载
jamasbabout 6 years ago
I&#x27;ve been doing some work on link prediction in knowledge graphs recently with poor results on real-world data. These methods don&#x27;t necessarily require a huge amount of data but they are very sensitive to noise and the &#x27;density&#x27; of dataset. The benchmark datasets are, in essence, very easy to get good performance on. It&#x27;s a real shame that metrics for these methods&#x27; tolerance of noise and sparsity are not reported because these are going to be present in almost any real-world dataset in far greater quantities than current benchmarks.
评论 #19581255 未加载
piloochabout 6 years ago
Looks very seriously made and documented, congrats ! Was looking at it a bit closely the other days and put it onto my list of future tools. There&#x27;s been another somewhat related library released by facebook recently, <a href="https:&#x2F;&#x2F;ai.facebook.com&#x2F;blog&#x2F;open-sourcing-pytorch-biggraph-for-faster-embeddings-of-extremely-large-graphs&#x2F;" rel="nofollow">https:&#x2F;&#x2F;ai.facebook.com&#x2F;blog&#x2F;open-sourcing-pytorch-biggraph-...</a>
评论 #19577515 未加载
bravuraabout 6 years ago
Can you help me understand, what are possible inputs to ampligraph?<p>I think the main use-case is plugging in an existing knowledge graph, and it filling in the gaps, correct?<p>Can I augment this will really high-quality embeddings for the nodes, that were learned over auxiliary unlabelled text?<p>What are other ways I can augment the data set?<p>Is this useful only when there are many edge-types, or is it also good when there are very few?<p>It looks promising, I just couldn&#x27;t immediately grok when I use should look to this library.
评论 #19580221 未加载
评论 #19581179 未加载
quenstionsaskedabout 6 years ago
Cool. KGE methods are becoming more and more useful as companies are trying to find ways to interface some internal knowledge graph with machine learning techniques. I expect this space to grow substantially!
评论 #19577554 未加载
mulletboyabout 6 years ago
btw, we are hiring research engineers here in our Dublin Lab. Send me an email if interested: luca.costabello@accenture.com <a href="https:&#x2F;&#x2F;www.accenture.com&#x2F;ie-en&#x2F;careers&#x2F;jobdetails?src=&amp;id=00693824_en" rel="nofollow">https:&#x2F;&#x2F;www.accenture.com&#x2F;ie-en&#x2F;careers&#x2F;jobdetails?src=&amp;id=0...</a>