[Xapian-devel] Doubt regarding Feature selection for 'Learning to Rank' algorithms

[Parth Gupta]

Hi Rishabh,

I had this feeling before. This is a really nice idea BUT we can not go
ahead with the project which is still not tested in the experimental
settings. Though it may be a wonderful research exercise, I would still
vote to go for the state-of-the-art methods which are completely published
with full details and experimental results. Hope you get my point.

Regards,
Parth.

On Wed, Apr 4, 2012 at 1:55 PM, Rishabh Mehrotra <erishabh at gmail.com> wrote:

> Hi Parth,
> Please find below the conversations I had with Hang Li sir and Prof. Liu
> regarding my proposed methodology.
> Regards,
>
> ---------- Forwarded message ----------
> From: Tie-Yan Liu <Tie-Yan.Liu at microsoft.com>
> Date: 2012/4/3
> Subject: RE: Doubt regarding Feature selection for 'Learning to Rank'
> algorithms
> To: "Hang Li (MSR)" <hangli at microsoft.com>, Rishabh Mehrotra <
> erishabh at gmail.com>
>
>
>  In my opinion, features definitely play an important role in learning to
> rank. Combining ListMLE with deep learning sounds interesting. We are eager
> to see the performance of your implementation. ****
>
> ** **
>
> Thanks****
>
> Tie-Yan****
>
> ** **
>
> *From:* Hang Li (MSR)
> *Sent:* 2012年4月1日 10:33
> *To:* Rishabh Mehrotra
> *Cc:* Tie-Yan Liu
> *Subject:* RE: Doubt regarding Feature selection for 'Learning to Rank'
> algorithms****
>
> ** **
>
> Rishabh****
>
> ** **
>
> Thank you for your interest.****
>
> ** **
>
> I am not quite sure whether it is easy to make improvement by combining
> deep learning and ListMLE. But you can try.  ****
>
> ** **
>
> To me, ListMLE is a very elegant model and since it is a log linear model,
> it appears to have a good match with deep learning.****
>
> ** **
>
> Maybe you can also get some comment from Tie-Yan.****
>
> ** **
>
> Hang****
>
> ** **
>
> *From:* Rishabh Mehrotra [mailto:erishabh at gmail.com]
> *Sent:* Saturday, March 31, 2012 1:03 AM
> *To:* Hang Li (MSR)
> *Subject:* Doubt regarding Feature selection for 'Learning to Rank'
> algorithms****
>
> ** **
>
> Hello sir,****
>
> ** **
>
> I attended your talk on Learning to Rank at MLSS 2011 at NUS Singapore
> last year in June. I was going through various Listwise approaches for
> ranking and ****
>
> the various features used to represent the documents.****
>
> ** **
>
> Recently Deep architectures have been used to learn feature
> representations in an unsupervised manner and have outperformed the
> state-of-the-art algorithms for various classification tasks. I am planning
> to work for an open-source organization and help them in implementing an
> algorithm for their Learing to Rank module. ****
>
> ** **
>
> I have decided to implement the *ListMLE algorithm* which you proposed in
> 2008. Instead of applying the conventional IR features I am thinking of
> using *Autoencoders* or other deep learning algorithm for feature
> extraction. It would really help me if you could comment on this decision:
> will it help me improve the performance or is it that features do not play
> that major a role in this sub-field of IR.****
>
> ** **
>
> Any suggestions form your side would really help me a lot. Thank you for
> your time.****
>
> ** **
>
> Best regards,****
>
> Rishabh Mehrotra,****
>
> BITS Pilani,****
>
> India.****
>
>
>
> --
> Rishabh.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.xapian.org/pipermail/xapian-devel/attachments/20120404/2421d293/attachment.htm>