Comments (3)
Hi @aiot-tech , the proposed method is not designed for incremental clustering, but I think it can be applied to incremental clustering with some adaptations.
- A naive solution is to merge the new data with the old ones and apply clustering again. It demands more computational cost but it can serve as a reference performance. (Actually, the performance of incremental clustering cannot surpass the full clustering, except there are other known priors, e.g., the new data does not share any classes with the old ones.)
- If it is known that the new data does not share any classes with the old ones, we can simply apply clustering on them separately and merge the results.
- If it is unknown whether there is overlap between new data classes and old data classes, the key lies in how to merge the new data into the old clusters. A possible solution is to decompose the problem into two stages, i.e., first apply clustering on them separately and then develop some techniques to merge clusters. For merging clusters, we can leverage the idea of super vertex to regard each cluster as a vertex for constructing an affinity graph, and then apply GCN-D and GCN-S thereon.
Besides extending the current full clustering approaches to incremental clustering, you can also design specialized algorithms for incremental clustering.
Above are some of my thoughts but I do not test them experimentally. Let me know if you make any progress.
from learn-to-cluster.
Great! Thanks a lot
from learn-to-cluster.
@aiot-tech Hi. Have you solved this problem? Can you share your idea?
from learn-to-cluster.
Related Issues (20)
- output error HOT 1
- Dataset issue HOT 2
- Issue while training LGCN HOT 1
- 您好,想请教一下训练GCN-D时报错run out of input HOT 1
- 没有修改任何参数跑作者提供的pretrained model for ms1m Fscore 只有0.32
- Clustering 512D features using pretrained model HOT 3
- building symmetric adjacency matrix in utils HOT 1
- why using output of relu for rebuilding knn graph HOT 1
- is not a checkpoint filemodels/pretrained_gcn_e_ms1m.pth HOT 3
- about extracting features HOT 3
- About how to make a .meta file HOT 1
- Can we manually set the number of clusters during the test HOT 1
- 大佬,请教下train_cluster_seg_iop_vox.sh 的test阶段报错的问题 HOT 2
- 两个face recognition模型,推荐使用哪个,hfsoftmax还要用到python 2 ? HOT 1
- About pretrained_model to get the the extracted feature HOT 2
- GCN-D MMdistributedDataparallel train HOT 2
- 关于性能 HOT 1
- How to apply lgcn for clustering on custom embeddings of collected face images? HOT 1
- On custom dataset preparation
- On the numpy version which may cause exception HOT 3
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 learn-to-cluster.