[Colloquium] REMINDER: 12/14 Machine Learning Seminar Series: Mickaël Binois, Argonne National Laboratory

[Mary Marre via Colloquium]

*When: *        Friday, December 14th, 11am-12pm
*Where:*        Room 526, TTIC, 6045 S Kenwood Avenue
*Who:    *       Mickaël Binois, Argonne National Laboratory

*Title: *          Improving Bayesian Optimization via Random Embeddings

*Abstract:* Bayesian optimization (BO) aims at efficiently optimizing
expensive black-box functions, such as hyperparameter tuning problems in
machine learning. Scaling up BO to many variables relies on structural
assumptions about the underlying black-box, to alleviate the curse of
dimensionality. In this talk, we review several options to this end, with
emphasis on the low effective dimensionality hypothesis. Starting with a
sampled random embedding, we discuss several practical issues related to
selecting a suitable search space. We also present alternative sampling
methods for the embedding, as well as techniques to identify the low
effective subspace. The performance and robustness gains of the proposed
enhancements are illustrated on numerical examples.

 For more information on the machine learning seminar series (MLSS), please
request to join the group at https://groups.google.com/a/ttic.edu/d/forum/
mlss. If you are interested in presenting in the seminar, please send an
email to steve at ttic.edu    <steve at ttic.edu>


Mary C. Marre
Administrative Assistant
*Toyota Technological Institute*
*6045 S. Kenwood Avenue*
*Room 517*
*Chicago, IL  60637*
*p:(773) 834-1757*
*f: (773) 357-6970*
*mmarre at ttic.edu <mmarre at ttic.edu>*


On Thu, Dec 13, 2018 at 11:15 AM Mary Marre <mmarre at ttic.edu> wrote:

> *When: *        Friday, December 14th, 11am-12pm
> *Where:*        Room 526, TTIC, 6045 S Kenwood Avenue
> *Who:    *       Mickaël Binois, Argonne National Laboratory
>
> *Title: *          Improving Bayesian Optimization via Random Embeddings
>
> *Abstract:* Bayesian optimization (BO) aims at efficiently optimizing
> expensive black-box functions, such as hyperparameter tuning problems in
> machine learning. Scaling up BO to many variables relies on structural
> assumptions about the underlying black-box, to alleviate the curse of
> dimensionality. In this talk, we review several options to this end, with
> emphasis on the low effective dimensionality hypothesis. Starting with a
> sampled random embedding, we discuss several practical issues related to
> selecting a suitable search space. We also present alternative sampling
> methods for the embedding, as well as techniques to identify the low
> effective subspace. The performance and robustness gains of the proposed
> enhancements are illustrated on numerical examples.
>
>  For more information on the machine learning seminar series (MLSS), please
> request to join the group at https://groups.google.com/a/ttic.edu/d/forum/
> mlss. If you are interested in presenting in the seminar, please send an
> email to steve at ttic.edu    <steve at ttic.edu>
>
>
>
>
>
> Mary C. Marre
> Administrative Assistant
> *Toyota Technological Institute*
> *6045 S. Kenwood Avenue*
> *Room 517*
> *Chicago, IL  60637*
> *p:(773) 834-1757*
> *f: (773) 357-6970*
> *mmarre at ttic.edu <mmarre at ttic.edu>*
>
>
> On Sat, Dec 8, 2018 at 10:55 PM Mary Marre <mmarre at ttic.edu> wrote:
>
>> *When:         *Friday, December 14th, 11am-12pm
>> *Where:        *Room 526, TTIC, 6045 S Kenwood Avenue
>> *Who:          * Mickaël Binois, Argonne National Laboratory
>>
>> *Title:*           Improving Bayesian Optimization via Random Embeddings
>>
>> *Abstract:* Bayesian optimization (BO) aims at efficiently optimizing
>> expensive black-box functions, such as hyperparameter tuning problems in
>> machine learning. Scaling up BO to many variables relies on structural
>> assumptions about the underlying black-box, to alleviate the curse of
>> dimensionality. In this talk, we review several options to this end, with
>> emphasis on the low effective dimensionality hypothesis. Starting with a
>> sampled random embedding, we discuss several practical issues related to
>> selecting a suitable search space. We also present alternative sampling
>> methods for the embedding, as well as techniques to identify the low
>> effective subspace. The performance and robustness gains of the proposed
>> enhancements are illustrated on numerical examples.
>>
>>
>> For more information on the machine learning seminar series (MLSS), please
>> request to join the group at
>> https://groups.google.com/a/ttic.edu/d/forum/mlss. If you are interested
>> in presenting in the seminar, please send an email to steve at ttic.edu
>> <steve at ttic.edu>
>>
>>
>>
>> Mary C. Marre
>> Administrative Assistant
>> *Toyota Technological Institute*
>> *6045 S. Kenwood Avenue*
>> *Room 517*
>> *Chicago, IL  60637*
>> *p:(773) 834-1757*
>> *f: (773) 357-6970*
>> *mmarre at ttic.edu <mmarre at ttic.edu>*
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20181214/d7e3523b/attachment-0001.html>