AI

ai@lists.wikimedia.org

263 discussions

Using AI to identify counterfeit images
by Pine W 26 Jun '18

26 Jun '18

This is in today's news: https://www.dpreview.com/news/5039410137/adobe-ai-spots-tampered-images-by-… Pine ( https://meta.wikimedia.org/wiki/User:Pine )

1 0

Idle GPU servers - 1080TI
by Pavel Machalek 06 Jun '18

06 Jun '18

Dear AI ers, we recently finished a large Tensorflow project and our two servers are sitting idle. Let me know if you want to rent machines with TEN GeForce GTX 1080TIs, and over 250GB RAM and 48 CPUs. Latest CUDA and NVIDIA drivers installed. Cheers Paul

2 1

Scoring platform status update
by Adam Wight 02 May '18

02 May '18

We've summarized our technical work over the last three months, please see the blog post for details: https://phabricator.wikimedia.org/phame/post/view/104/status_update_may_2_2… Our novel, collaborative auditing system "judgment and dialogue engine" (JADE) is deployed to the beta cluster and is now available for tool developers to explore and integrate with. Read more here, https://mediawiki.org/wiki/JADE A dynamically updated table of ORES support is now available for anyone tracking our progress: https://tools.wmflabs.org/ores-support-checklist/ Regards, Adam

1 0

Study on music habits
by Sylvain Castagnos 26 Apr '18

26 Apr '18

*** Apologies for cross postings *** Hello, We are researchers at the LORIA laboratory (Nancy, France) and we are conducting a study on musical listening habits of the population over the world. Our goal is to better understand the preferences, expectations, and needs of users as regards music. This questionnaire is divided into 3 sections and its duration is estimated at 5 minutes. Could you help us by completing this form and spreading it widely? Fill out form <https://docs.google.com/forms/d/e/1FAIpQLSeWqSXQh0D872L6cfCEcHhB1ojyBUntsrz…> For any questions related to this experience, please contact us at the following email address: etude-kiwi(a)loria.fr. The answers to this questionnaire are collected anonymously. Thank you for your participation ! Best regards, -- Sylvain CASTAGNOS KIWI Team LORIA Laboratory

1 0

In the news: "Your next phone may have an ARM machine learning processor"
by Pine W 18 Feb '18

18 Feb '18

In case you're interested: https://techcrunch.com/2018/02/13/your-next-phone-may- have-an-arm-machine-learning-processor/ Pine <https://meta.wikimedia.org/wiki/User:Pine> <https://en.wikipedia.org/wiki/User:CatherineMunro/Bright_Places>

1 0

Re: [AI] [Wikimedia-l] Siamese networks and image classification
by Pine W 31 Jan '18

31 Jan '18

Hi John, I think that these questions might be best asked on the AI mailing list, so I'm copying this thread to that list. I would suggest that further discussion should take place on the AI mailing list, because that's where the WMF experts on AI subjects are likely to participate. Thanks for your interest in these questions. I too would be interested in automating image categorization for Commons. I think that full automation of image categorization would be a long term project, perhaps over the course of decades, but there may be ways to make incremental progress. :) Pine <https://meta.wikimedia.org/wiki/User:Pine> <https://en.wikipedia.org/wiki/User:CatherineMunro/Bright_Places> On Wed, Jan 24, 2018 at 1:20 PM, John Erling Blad <jeblad(a)gmail.com> wrote: > I had a plan to do a more thorough description on meta, but my plans are > always to optimistic… ;) > > Question 1: I wonder if anyone has done any work on categorization of > images on Commons by using neural nets. > Question 2: I wonder if anyone has used Siamese networks to do more general > categorization of images. > > A Siamese network is a kind of network used for face recognition. It is > called Siamese network because it is two parallel networks in its simplest > form, but one of the networks are precomputed so in effect you only > evaluates one of the networks. Usually you have one face provided by a > camera and run through one network, and a number of other precomputed faces > for the other network. When the difference is small, then you have a face > match. > > In general you can have any kind of image as long as you can train a kind > of fingerprint type of vector. This fingerprint will then acts like > locality-sensitive hashing [1]. Because you can create this LSH from the > vector, you can store alternate models in database, so you can later search > it and find usable models. Those models can then interpret the vector form > of the fingerprint, and by evaluating those models with actual input (ie > the fingerprint vector) you can get an estimate for how probable it is that > the image belong to a specific category. > > What you would get from the previous process is a list of probable > categorizations. That list can be sorted, and a user trying to categorize > an image can then chose to select some of the categories. > > Note that the output of the network is not just a few categories, it can be > a variant number of models that outputs a probability for a specific > category. It is a bit like a map-reduce, where the you find (filter) the > possible models and then evaluate (map) those models with the fingerprint > vector. > > It is perhaps not obvious, but the idea behind Siamese networks are two > parallel networks, but the implementation is with a single active network. > Also the implemented network has a first part that computes the fingerprint > vector (kind of a bottleneck network) and the second part are the stored > models that takes this fingerprint vector and calculates a single output > for the probability of a single category. > > [1] https://en.wikipedia.org/wiki/Locality-sensitive_hashing > > On Mon, Jan 22, 2018 at 6:47 AM, Pine W <wiki.pine(a)gmail.com> wrote: > > > Hi John, > > > > I am having a little trouble with understanding your email from January > > 15th. Could you perhaps state your question or point in a different way? > > > > Pine <https://meta.wikimedia.org/wiki/User:Pine> > > <https://en.wikipedia.org/wiki/User:CatherineMunro/Bright_Places> > > > > On Mon, Jan 15, 2018 at 5:55 AM, John Erling Blad <jeblad(a)gmail.com> > > wrote: > > > > > This is the same as the entry on the wishlist for 2016, but describes > the > > > actual method. > > > https://meta.wikimedia.org/wiki/2016_Community_Wishlist_ > > > Survey/Categories/Commons#Use_computer_vision_to_propose_categories > > > > > > Both contrastive and triplet loss can be used while learning, but > neither > > > are described at Wikipedia. > > > > > > On Sun, Jan 14, 2018 at 8:16 PM, Pine W <wiki.pine(a)gmail.com> wrote: > > > > > > > Hi John, > > > > > > > > I have not heard of an initiative to use Siamese neural networks for > > > image > > > > classifications on on Commons. You might make a suggestion on the AI, > > > > Research, and/or Commons mailing lists regarding this idea. You might > > > also > > > > make a suggestion in IdeaLab > > > > <https://meta.wikimedia.org/wiki/Grants:IdeaLab>. > > > > > > > > Pine <https://meta.wikimedia.org/wiki/User:Pine> > > > > <https://en.wikipedia.org/wiki/User:CatherineMunro/Bright_Places> > > > > > > > > On Sun, Jan 14, 2018 at 3:46 AM, John Erling Blad <jeblad(a)gmail.com> > > > > wrote: > > > > > > > > > Has anyone tried to use a Siamese neural network for image > > > classification > > > > > at Commons? I don't know if it will be good enough to run in > > autonomous > > > > > mode, but it will probably be a huge help for those that do manual > > > > > classification. > > > > > > > > > > Imagine a network providing a list of possible categories, and the > > user > > > > > just ticks off usable categories. > > > > > > > > > > A Siamese network can be learned by using a triplet loss function, > > > where > > > > > the anchor and the positive candidate comes from the same category, > > and > > > > the > > > > > negative candidate comes from an other category but are otherwise > > close > > > > to > > > > > the anchor. > > > > > > > > > > Output from the network is like a fingerprint, and those > fingerprints > > > can > > > > > be compared to other images with known fingerprints, or against a > > > > > generalized fingerprint for a category. > > > > > > > > > > John Erling Blad >

1 0

Scoring platform team status update
by Adam Wight 30 Jan '18

30 Jan '18

Friends, Anyone interested in ORES and Scoring Platform’s recent progress can read an exhaustive list on our engineering blog, https://phabricator.wikimedia.org/phame/post/view/84/status_update_january_… <https://phabricator.wikimedia.org/phame/post/view/84/status_update_january_…> Aside from supporting several new languages, the biggest changes that might affect other developers are that: we now supply enwiki new article draft quality predictions, already loaded into the MediaWiki database via Extension:ORES; and all of our models now include statistics that can be used to calculate appropriate cutoff points. Enjoy! -Amir, Aaron, Adam, and Sumit

1 0

Research Project on Real-Time Vandal Detection during Editing
by Martin Potthast 20 Jan '18

20 Jan '18

Hi everyone, we [1] would like to announce a research project with the goal of studying whether user interactions recorded at the time of editing are suitable to predict vandalism in real time. Should vandal editing behavior be sufficiently different from normal editing behavior, this would allow for a number of interesting real-time prevention techniques. For example: - withholding confidently suspicious edits for review before publishing them, - a popup asking "I am not a vandal" (as in Google's "I am not a robot") to analyze vandal reactions, - a popup with a chat box to personally engage vandals, e.g., to help them find other ways of stress relief or to understand them better, - or at the very least: a new signal to improve traditional vandalism detectors. We have set up a laboratory environment to study editor behavior in a realistic setting using a private mirror of Wikipedia. No editing whatsoever is conducted on the real Wikipedia as part of our experiments, and all test subjects of our user studies are made aware of the experimental nature of their editing. We plan on making use of crowdsourcing as a means to attain scale and diversity. If you wish to participate in this study as a test subject yourself, please get in touch. The more diversity, the more insightful the results will be. We are also happy to collaborate and to answer all questions that may arise in relation to the project. For example, our setup and tooling may turn out to be useful to study other user behavior-related things without having to actually deploy experiments within the live MediaWiki. Best, Martin PS: The AICaptcha project seems most closely related. @Vinitha and Gergő: If you wish, we can set up a Skype meeting to talk about a avenues for collaboration. [1] A group of students and researchers from Bauhaus-Universität Weimar ( www.webis.de) and Leipzig University (www.temir.org); project PI: Martin Potthast.

1 0

Improving Captcha on account creation page using Machine Learning
by Vinitha Vs 04 Jan '18

04 Jan '18

Hi, I[0] am an Outreachy Round 15 intern and I am working on AICaptcha[1] project. This project is aimed at creating a better captcha system (like the Google invisible captcha) which can prevent/reduce the incidence of bots creating user accounts and spamming Wikipedia. My mentors on this project are Gergő Tisza[2] and Adam Roses Wight[3]. The key aspects of this project are: 1. Data capture for training a machine learning classifier which is elaborated in Phabricator task[4]. The data can be captured from the registration page using the WikiMediaEvents extension. 2. Feature selection, dealing with selecting the most appropriate features which can improve the classification model, explained in Phabricator task[5] 3. Finding appropriate machine learning classifier to create the model [6]. Kindly provide suggestions/ideas on Phabricator, so that any idea missed by oversight can be discussed. Also if there are any possible issues which I have not thought about yet, please comment on the tasks so that I can take care of them sooner rather than later. Links: [0]: https://meta.wikimedia.org/wiki/User:Groovier [1] : https://phabricator.wikimedia.org/project/profile/3137/ [2]: https://www.mediawiki.org/wiki/User:Tgr_(WMF) [3]: https://www.mediawiki.org/wiki/User:Adamw [4]: https://phabricator.wikimedia.org/T183991 [5]:https://phabricator.wikimedia.org/T183998 [6]: https://phabricator.wikimedia.org/T184013 Thanks Vinitha

1 0

CFP for IEEE ICCC 2018
by Incheon Paik 18 Dec '17

18 Dec '17

Apology for cross-postings. Welcome your submission, and please send it to interested colleagues and students. Thanks! Incheon ***** Call For Papers ***** The Second IEEE International Conference on Cognitive Computing (IEEE ICCC 2018) 2018 IEEE World Congress on Services JULY 2 - 7, 2018, SAN FRANCISCO, USA http://conferences.computer.org/iccc/2018/cfp/ The 2018 IEEE International Conference on Cognitive Computing (ICCC 2018) focuses on the integration of cognitive computing with service computing for providing intelligence as a service. Technical topics of interest include but are not limited to the following: I. Cognitive Computing Technologies and Infrastructure - Cognitive Computing Fundamentals such as Mind, Language, Perception, Memory, Attention, Reasoning, Emotion, Prediction, and Problem Solving - Cognitive Computing Techniques using Artificial Intelligence, Pattern and Speech Recognition, and Natural Language Processing - Cognitive Approaches using Data Mining, Machine Learning, and Deep Learning Techniques - Consciousness and Awareness - Cognitive Data Analytics - Systems for Brain - Machine Interface - Collective Intelligence and Wisdom - Semantics for Conversation and Pragmatics - Service Science for Cognitive Computing II. Cognitive Computing Systems and Applications - Deep Learning for Cognition - Brain Analysis for Cognition - Big Data Infrastructure for Cognition - Distributed Processing for Cognition - Consciousness and Awareness Framework - Cognitive Agents and Robots - Semantics for Conversation and Pragmatics - Service Technologies for Cognitive Computing - Cyber cognitive systems and services - Internet of Cognitive Things - Big Data Analytics for Cognition III. Intelligence of Innovation Lifecycle and Sensing - Intelligence Innovation Lifecycle (Such as Intelligence Producing, Extracting, Understanding, Perceiving, Responding, Inferencing, Correlating, Propagating, Billing, Exchanging, Trading, Replicating, Destroying, and Repairing) - Cognitive Sensing (Sight, Hearing, Taste, Smell, and Touch) - Cognitive Data (Image, Sound, Text, Web Data, Sensor Data, and Machine Generated Data) - Cognitive Environment (Kinesthesia, Balance, Acceleration, and Internal Stimuli) *** Manuscript Guidelines and Submission Information *** Language: English. Paper size and format: US Letter; Two-column format in the IEEE style. Page limit: Up to 8 pages for peer review. Abstract length: Approximately 1500 characters. File format: Limit the size of a single PDF file to be 6 MB. All submitted manuscripts will be peer-reviewed by at least 3 program committee members. Accepted papers with confirmed presentation will appear in the conference proceedings published by the IEEE Computer Society Press. Each conference/congress will publicly announce the winners of its Best Paper Award, and Best Student Paper Award. The authors of selected papers will be encouraged to submit extended and enhanced versions of their papers to the IEEE Transactions on Services Computing (TSC) and other suitable journals. *** Key Dates *** Paper submission (for peer review) due: January 10, 2018. Notification to authors on the decisions on papers: March 15, 2018. Camera-ready papers due: April 6, 2018. Congress date: July 2-7, 2018. ICCC 2018 General Chairs ======================== Jeffrey J.P. Tsai, Asia University, Taiwan Ernesto Damiani, University of Milan, Italy ICCC 2018 Program Chairs ======================== Xiaoqing Frank Liu, University of Arkansas, USA Incheon Paik, University of Aizu, Japan SERVICES Congress General Chair: Stephen S. Yau, Arizona State University, USA IEEE-CS Technical Committee on Services Computing Chair: Rong N. Chang, IBM Research, USA Steering Committee: Elisa Bertino, Purdue University, USA Carl K. Chang, Iowa State University, USA (Chair) Rong N. Chang, IBM Research, USA Peter Chen, Carnegie Mellon University. USA Ernesto Damiani, University of Milan, Italy Ian Foster, University of Chicago, USA Dennis Gannon, Indiana University, USA Michael Goul, Arizona State University, USA Frank Leymann, University of Stuttgart, Germany Hong Mei, Beijing Institute of Technology, China Stephen S. Yau, Arizona State University, USA Finance Chair: Shiyong Lu, Wayne State University, USA Publication Chair: Katsunori Oyama, Nihon University, Japan Publicity Team: Patrick C. K. Hung, University of Ontario Institute of Technology, Canada Eleanna Kafeza, Zayed University, UAE Hamad Binsalleeh, Al-Imam Muhammad Ibn Saud Islamic University, Saudi Arabia Jorge Roa, National Technological University, Santa Fe Regional Faculty, Argentina Marcelo Fantinato, University of Sao Paulo, Brazil Young Yoon, Hongik University, Korea Radha Krishna Pisipati, Infosys, India Uwe Breitenb點her, University of Stuttgart, Germany Andrei V. Kopylov, Tula State University, Russia Matogoro Jabera, University of Dodoma, Tanzania Mohammad Aazam, Carnegie Mellon University, Qatar Bhekisipho Twala, University of South Africa, South Africa If you have any questions or queries on IEEE ICCC 2018, please send email to: ieeecs DOT iccc AT gmail DOT com.

1 0

← Newer
1
...
10
11
12
13
14
15
16
...
27
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

AI