Posted on

Vowpal Wabbit 7.8 at NIPS

I just created Vowpal Wabbit 7.8, and we are planning to have an increasingly less heretical followup tutorial during the non-“ski break” at the NIPS Optimization workshop. Please join us if interested.

I always feel like things are going slow, but in the last year, but there have been many changes overall. Notes for 7.7 are here. Since then, there are several areas of improvement as well as generalized bug fixes and refactoring.

  1. Learning to Search: Hal completely rewrote the learning to search system, enough that the numbers here are looking obsolete. Kai-Wei has also created several advanced applications for entity-relation and dependency parsing which are promising.
  2. Languages Hal also created a good python library, which includes call-backs for learning to search. You can now develop advanced structured prediction solutions in a nice language. Jonathan Morra also contributed an initial Java interface.
  3. Exploration The contextual bandit subsystem now allows evaluation of an arbitrary policy, and an exploration library is now factored out into an independent library (principally by Luong with help from Sid and Sarah). This is critical for real applications because randomization must happen at the point of decision.
  4. Reductions The learning reductions subsystem has continued to mature, although the perfectionist in me is still dissatisfied. As a consequence, it’s now pretty easy to program new reductions, and the efficiency of these reductions has generally improved. Several news ones are cooking.
  5. Online Learning Alekh added an online SVM implementation based on LaSVM. This is known to parallelize well via the para-active approach.

This project has grown quite a bit—there are about 30 different people contributing to VW since the last release, and there is now a VW meetup (December 8th!) in the bay area that I wish I could attend.

Posted on

An ICML proposal: yearly surveys

I’d like to propose that ICML conducts a yearly survey similar to the one from 2010 or 2012 which is reported to all.

The key reason for this is information: I expect everyone participating in ICML has some baseline interest in how ICML is doing. Everyone involved has personal anecdotal information, but we all understand that a few examples can be highly misleading.

Aside from satisfying everyone’s joint curiousity, I believe this could improve ICML itself. Consider for example reviewing. Every program chair comes in with ideas for how to make reviewing better. Some succeed, but nearly all are forgotten by the next round of program chairs. Making survey information available will help quantify success and correlate it with design decisions.

The key question to ask for this is “who?” The reason why surveys don’t happen more often is that it has been the responsibility of program chairs who are typically badly overloaded. I believe we should address this by shifting the responsibility to a multiyear position, similar to or the same as a webmaster. This may imply a small cost to the community (<$1/participant) for someone's time to do and record the survey, but I believe it's a worthwhile cost. I plan to bring this up with IMLS board in Beijing, but would like to invite any comments or thoughts.

Posted on

The New York ML Symposium, take 2

The 201314 is New York Machine Learning Symposium is finally happening on March 28th at the New York Academy of Science. Every invited speaker interests me personally. They are:

We’ve been somewhat disorganized in advertising this. As a consequence, anyone who has not submitted an abstract but would like to do so may send one directly to me (jl@hunch.net title NYASMLS) by Friday March 14. I will forward them to the rest of the committee for consideration.

Posted on

NIPS tutorials and Vowpal Wabbit 7.4

At NIPS I’m giving a tutorial on Learning to Interact. In essence this is about dealing with causality in a contextual bandit framework. Relative to previous tutorials, I’ll be covering several new results that changed my understanding of the nature of the problem. Note that Judea Pearl and Elias Bareinboim have a tutorial on causality. This might appear similar, but is quite different in practice. Pearl and Bareinboim’s tutorial will be about the general concepts while mine will be about total mastery of the simplest nontrivial case, including code. Luckily, they have the right order. I recommend going to both 🙂

I also just released version 7.4 of Vowpal Wabbit. When I was a frustrated learning theorist, I did not understand why people were not using learning reductions to solve problems. I’ve been slowly discovering why with VW, and addressing the issues. One of the issues is that machine learning itself was not automatic enough, while another is that creating a very low overhead process for doing learning reductions is vitally important. These have been addressed well enough that we are starting to see compelling results. Various changes:

  • The internal learning reduction interface has been substantially improved. It’s now pretty easy to write new learning reduction. binary.cc provides a good example. This is a very simple reduction which just binarizes the prediction. More improvements are coming, but this is good enough that other people have started contributing reductions.
  • Zhen Qin had a very productive internship with Vaclav Petricek at eharmony resulting in several systemic modifications and some new reductions, including:
    1. A direct hash inversion implementation for use in debugging.
    2. A holdout system which takes over for progressive validation when multiple passes over data are used. This keeps the printouts ‘honest’.
    3. An online bootstrap mechanism system which efficiently provides some understanding of prediction variations and which can sometimes effectively trade computational time for increased accuracy via ensembling. This will be discussed at the biglearn workshop at NIPS.
    4. A top-k reduction which chooses the top-k of any set of base instances.
  • Hal Daume has a new implementation of Searn (and Dagger, the codes are unified) which makes structured prediction solutions far more natural. He has optimized this quite thoroughly (exercising the reduction stack in the process), resulting in this pretty graph.
    part of speech tagging time accuracy tradeoffs
    Here, CRF++ is commonly used conditional random field code, SVMstruct is an SVM-style approach to classification, and CRF SGD is an online learning CRF approach. All of these methods use the same features. Fully optimized code is typically rough, but this one is less than 100 lines.

I’m trying to put together a tutorial on these things at NIPS during the workshop break on the 9th and will add details as that resolves for those interested enough to skip out on skiing 🙂

Edit: The VW tutorial will take place during the break at the big learning workshop from 1:30pm – 3pm at Harveys Emerald Bay B.