Comments (7)
I wanted to quickly follow up on our separate email conversation: We identified
the issue and are finishing additional experiments. I will update this issue in
the next days with the results. Thanks again for filing this!
from poincare-embeddings.
Correction in the hyperparameters description above. We are using:
-dim 200
not "dims" ...
from poincare-embeddings.
I am still seeing this. Is there any update? Would be very important for everyone to know what's going on. Are the results presented in the NIPS'17 paper wrong/misleading?
from poincare-embeddings.
Same situation here, wondering if there's any update on this?
from poincare-embeddings.
Thank you for raising this again and sorry for the delay. In addition to our follow-ups over email we should have updated this issue on Github as well.
Basically: The reason for the stronger Euclidean baseline using the open sourced code is that the paper used a different setting where the Euclidean embeddings were regularized (similar to previous work). With open sourcing the code we disabled this regularization by default and it turned out to work better (as pointed out by Sameer). Since it led to a stronger Euclidean baseline in higher dimensions we decided to keep it like that in the code. Hyperbolic embeddings provide a substantial performance improvement in lower dimensions, which is really the main focus of this work.
from poincare-embeddings.
Thank you for raising this again and sorry for the delay. In addition to our follow-ups over email we should have updated this issue on Github as well.
Basically: The reason for the stronger Euclidean baseline using the open sourced code is that the paper used a different setting where the Euclidean embeddings were regularized (similar to previous work). With open sourcing the code we disabled this regularization by default and it turned out to work better (as pointed out by Sameer). Since it led to a stronger Euclidean baseline in higher dimensions we decided to keep it like that in the code. Hyperbolic embeddings provide a substantial performance improvement in lower dimensions, which is really the main focus of this work.
Sorry, I'm not sure I fully understand how the Euclidean embeddings were regularized... could you please explain a little bit more? A lot of thanks.
from poincare-embeddings.
@martinthewhale I think the idea is that since there is exponentially more "space" near the boundary of the Poincare ball, the easiest way for the algorithm to minimise the loss is to push all nodes outwards. This is a form of overfitting since we ideally want nodes that are higher in the original hierarchy to be kept closer to the centre of the ball. I believe this is achieved by regularising the norm of v in equation 6 in the paper. This means that for every parent(v)/child(u) relationship we consider we are always encouraging parents (nodes higher in the hierarchy) to stay closer to the origin.
from poincare-embeddings.
Related Issues (20)
- The distance function of euclidean manifold is wrong.
- Error when running NIPS 2017 Release HOT 2
- What is the output after learning? HOT 2
- [Question] Running Inference on New Nodes HOT 1
- euclidean embedding HOT 3
- [Question] Predicting the parent of an unseen word in an existing hierarchy HOT 4
- Restore from checkpoints and train?
- NIPS results not reproducible with this code.
- ValueError: Buffer dtype mismatch, expected 'long_t' but got 'long' HOT 1
- Why the Euclidean gradient omit the first part ? HOT 2
- stop when eval HOT 2
- How to get embeddings? HOT 1
- Hyperparameters for reproduction HOT 1
- mammal_closure.csv not found HOT 1
- Entailment cones compute the wrong angle? HOT 3
- What should we do after training? HOT 4
- Is there any filter file for plants and vehicles subtrees like 'mammals_filter.txt'? Could you please share these files?
- how was mammal_closure.csv created HOT 1
- Large dataset that needs continuous training HOT 1
- Problems related to constrain the embeddings to remain within the Poincaré ball via the projection
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from poincare-embeddings.