mrintellectual
Joined 12 years ago
Karma
956
Submissions
138
Member since
Oct 20
Recent submissions
Comment · 4 years ago
Although a lot of folks are undoubtedly disappointed, props to Magnus for understanding that there are other great accomplishments to be had besides continuously winning the WCC. I think the format of the championship match was a deal breaker for him - months of preparation and a slew of classical games meant that he would have little time to devote to other shorter time format tournaments.
With that being said, match between Ian and Ding would also be incredibly entertaining. I look forward to it.
Comment · 4 years ago
I remember this from a while back - good to bring back memories. This may be a bit excessive, but using CLIP (https://towhee.io/image-text-embedding/clip) or some other multimodal learning model in conjunction with Milvus (https://milvus.io) could make for a search engine that takes the graphics into consideration as well. Not sure how well it would work for comics, but it could make for a pretty unique weekend hack.
Comment · 4 years ago
Good catch on the disclosure, I edited my original comment to reflect this fact.
On the topic of vector search, Milvus is another great vector database - it's open source and we provide single-line startup scripts via `docker-compose` in addition to installation via apt & yum (https://milvus.io/docs/install_standalone-docker.md). There are also no restrictions on the number of vectors that users can store. Internally, we've successfully scaled Milvus to handle billion+ vectors, while many of our users have stored hundreds of millions of vectors in a production environments as well.
Comment · 4 years ago
Great article. We used something very similar to help implement simlarity search at Yahoo a couple years back (https://yahooresearch.tumblr.com/post/158115871236/introduci...). We were using a indexing strategy called Locally Optimized Product Quantization, which worked great in terms of query times but required a training procedure which made successive inserts fairly inefficient.
Thankfully, we have a much wider variety of indexing options these days (https://milvus.io/docs/index.md) in addition to powerful vector databases (https://zilliz.com/learn/what-is-vector-database). I'm glad to see the barrier to entry for semantic image retrieval becoming lower and lower as ML infrastructure matures.
[EDIT] Disclosure: I work at Zilliz.
Comment · 4 years ago
Great article. I've been working in and around this space since 2014, and I think similarity learning, vector search, and embedding management will be a core part of future applications that leverage ML.
I recently built a similarity search application that recommends new Pinterest users channels to follow based on liked images using Milvus (https://github.com/milvus-io/milvus) as a backend. Similarity learning is a huge part of it, and I'm glad more and more tools like Quaterion are being released to help make this kind of tech ubiquitous.
Comment · 4 years ago
The list of projects seems to be at least partially based on https://github.com/RunaCapital/awesome-oss-alternatives.
Website: https://frankzliu.com
Github: https://github.com/fzliu
Twitter: @frankzliu