Whoa! Lewis, can you give some more detail on this talk, what you proposed, and what you plan to talk about?
I haven't ever been to ApacheCon, but am interested in going. I don't have much of a feel for what motivates folks outside the academic research community, and that would be good to have in laying out projects that might interest people. Regarding those project, I have a number of them. Perhaps it would be useful to flesh them out with some more detail, and perhaps post them, for those who are interested. First, with respect to Tommaso's question, the following: - Use cases. I'd really like to push machine translation as a black box, where people can download and use models, not caring how they work, and building on top of them. I think this could be transformative. I've just added to Joshua the ability to add, store, and manage custom phrasal translation rules, which would let people take a model and add their own translations on top of it, perhaps correcting mistakes as they encounter them. There's a JSON API for it (undocumented). Building this up would also require pulling together lots of different test sets, evaluating changes, and so on. - Neural nets. This is a huge research area. I think the advantages are that it could enable releasing models that are much smaller. However, on the down side, it's not clear what the best way to integrate these models into Joshua is. Fully neural attention models would require re-architecting Joshua, as they are essentially a new paradigm. Adding neural components as feature functions that interact with the existing decoding algorithm would be an intermediate step. For other projects, I'd love: - Better documentation, developer and end-user (probably I need to write a lot of this; if nothing else, it would be hugely useful to me in terms of prioritizing to know that people want it) - Rewriting certain components. The tuning modules, in particular, are a real mess, and should be synthesized and improved. - Replacing Moses components. Joshua can call out to Moses to build phrase tables; it would be nice to get rid of this (and wouldn't be that hard) with our own Java implementations. It would also be good to add a lexicalized distortion model to the phrase-based decoder. matt > On Mar 14, 2016, at 9:37 AM, Lewis John Mcgibbney <lewis.mcgibb...@gmail.com> > wrote: > > Hi Kellen, > I'll be there for sure. I actually had a talk accepted which is an entire > overview of Joshua and will probably touch on what the roadmap is through > incubation and beyond. > In short I look forward to tagging up with you and any others are Ho will > be there/we can interest in the project. > A side note here is that more projects and talks at ApacheCon are becoming > increasingly relevant for science and healthcare so I hope we can do a real > justice to Joshua by undertaking a driven community building effort > @ApacheCon. > This being said, let's hash out here some objectives, must haves, would > likes, roadmap, etc if you like. > Thanks for brining this topic up. > Lewis > > On Monday, March 14, 2016, Tommaso Teofili <tommaso.teof...@gmail.com> > wrote: > >> Hi Kellen, >> >> I won't be at ApacheCon unfortunately, however your idea sounds good if >> there's anyone else joining. >> Perhaps we can also discuss some of the topics you mention here on the >> list, and maybe that would be of help for f2f chats too. >> >> I'd particularly interested in discussing: >> - use cases >> - ideas around usage of neural networks for MT >> >> Regards, >> Tommaso >> >> >> >> >> Il giorno lun 14 mar 2016 alle ore 12:30 kellen sunderland < >> kellen.sunderl...@gmail.com <javascript:;>> ha scritto: >> >>> Hello Joshua fans, >>> >>> I was wondering how many Joshua/Apache folks are going to be at ApacheCon >>> in May. I was hoping if there's enough of us we could get together and >>> chat about development ideas for Joshua, discuss each others use cases, >>> etc. I know that I for one would love to sync up on how the development >>> model for the project will work in the future. Maybe ApacheCon would be >> a >>> good opportunity to chat face to face. >>> >>> -Kellen Sunderland >>> >> > > > -- > *Lewis*
