06:36:35 RRSAgent has joined #computationalintelligence 06:36:39 logging to https://www.w3.org/2023/09/13-computationalintelligence-irc 06:36:39 RRSAgent, do not leave 06:36:43 RRSAgent, make logs public 06:36:44 Meeting: Unleashing the Power of Computational Intelligence on the Web 06:36:44 Chair: Rachel Yager, Gabriella Pasi 06:36:45 Agenda: https://github.com/w3c/tpac2023-breakouts/issues/24 06:36:49 clear agenda 06:36:51 agenda+ Pick a scribe 06:36:53 agenda+ Reminders: code of conduct, health policies, recorded session policy 06:36:55 agenda+ Goal of this session 06:36:57 agenda+ Discussion 06:36:59 agenda+ Next steps / where discussion continues 07:20:14 tidoust has joined #computationalintelligence 07:21:57 Rachel has joined #computationalintelligence 07:24:10 AdamSobieski has joined #computationalintelligence 07:24:10 agenda? 07:24:10 JohnRochford has joined #computationalintelligence 07:24:10 present+ 07:24:10 Joshua_Lochner has joined #computationalintelligence 07:25:46 Maud has joined #computationalintelligence 07:28:01 dsinger has joined #computationalintelligence 07:30:08 agenda item 1 Session Opening by Rachel Yager 07:30:08 agenda? 07:32:15 agenda 1 = Session Opening by Rachel Yager 07:32:15 JohnRochford_ has joined #computationalintelligence 07:32:15 agenda 2 = Rachel Yager - Community Roadmap 07:32:15 scribe+ 07:32:51 agenda 3 = Gabriella Pasi - CI Overview 07:33:08 agenda 4 = Joshua Lochner - Using Hugging Face in your JavaScript applications 07:33:52 agenda 5 = Adam Sobieski - Towards a Protocol for Computational Intelligence as a Service 07:33:52 agenda 6 = Luigi Troiano - AI Standardization Experience 07:33:53 agenda+ Victoria Lopez - Computational Intelligence & Health 07:34:09 agenda 6 = Luigi Troiano - AI Standardization Experience 07:34:37 agenda 7 = agenda 6 = Victoria Lopez - Computational Intelligence & Health 07:35:10 agenda? 07:35:31 agenda+ Victoria Lopez - Computational Intelligence & Health 07:35:37 agenda? 07:35:58 zkis has joined #computationalintelligence 07:36:07 present+ Zoltan_Kis 07:36:50 present+ 07:36:54 fremy has joined #computationalintelligence 07:42:23 present+ 07:43:11 take up item 1 07:45:46 q? 07:45:49 q+ 07:46:05 https://www.w3.org/community/blog/2023/08/12/proposed-group-computational-intelligence-business-group/ 07:46:45 q- 07:47:02 https://www.w3.org/groups/wg/webmachinelearning/ 07:47:40 scribe+ 07:48:27 Difficult to define unanimously, first definition by IEEE 07:49:19 1990 and creation of computational Intelligence (CI) society in 2003 07:49:49 take up item 2 07:49:58 3 pillars neural networks, fuzzy systems and evolutionary computations 07:50:10 take up item 3 07:50:12 Vagner_NIC_br has joined #computationalintelligence 07:50:50 Jim Bezdek subset of AI 07:51:07 relies on soft computing methods 07:51:57 nick Cercone 2022 web intelligence 07:52:25 Applications: web mining, semantic web, Natural Language processing 07:53:18 fake news identification, chatbot classification,... 07:54:38 take up item 4 07:54:47 Using Hugging face JS in your applications 07:54:53 transformers and more 07:55:40 Joshua's bridges gap between web dev and Machine learning 07:56:14 hugging face AI community 07:56:24 love open source (libraries) 07:57:03 JavaScriptification of AI 07:57:58 modules Inference (calls ), hubs users interact with hugging face hub, agents library 07:58:33 28 different tasks in inference speech recognition, translation for free 07:58:50 text to image, 07:59:37 create models and files to repositories,... 08:00:17 Agent JS you can ask the LLM draw me a picture, it will generate and run the code (stable fusion for cat) 08:01:05 transformers JS 6000 downloads (high use) 08:02:13 supported tasks: text classification, code completion, text-to-text 08:03:19 image classification, object detection, segmentation (text, vision, audio, multimodal) 08:04:17 speech-to-text (open AI model) 08:04:36 transcribe audio file whisper web, real time 08:04:48 text-to speech in the future 08:04:56 embeddings, ... 08:05:27 Applications: what can you do with tools: talk to a youtube video 08:05:40 browser extension 08:05:51 semantic image search engine 08:06:16 search using natural language across a DB 08:06:26 doodle dash 08:07:43 take up item 5 08:07:56 Protocol for CI as a service 08:08:37 use cases telerobotics, videogame AI, 08:09:07 useful for robotics , videogames,... 08:09:19 simulation training AI systems 08:09:46 evaluation of AI system 08:10:16 discussion sensors control signals 08:10:27 Time synchronization 08:10:35 remote procedure calls 08:10:42 robot to robot communication 08:10:57 extensibility (new types of sensor data..) 08:11:18 Smatnic (future opportunities) 08:12:17 Conclusion control robots and avatars, existing protocols can be extended ... 08:12:22 take up item 6 08:12:25 Adam's presentation link: https://github.com/w3c/tpac2023-breakouts/files/12591451/Adam.Sobieski.Towards.a.Protocol.for.Computational.Intelligence.as.a.Service.pdf 08:13:00 Standardization in AI/CI (ISO level) 08:13:09 Interoperability 08:13:58 to make models work together at syntactic and semantic level 08:14:12 Trustworthiness legal implication 08:14:38 EU governments require compliance 08:15:14 Way data are managed is important 08:15:23 consider legal implications 08:15:32 Internal mechanics 08:15:45 models spread over different components 08:15:55 how to make communication efficient 08:16:27 propagating learning experience over nodes 08:16:48 Metrics 08:17:02 Measuring the way we use these resources 08:17:17 competition between models 08:17:26 different ways to look at performance 08:17:37 Should be shared 08:17:44 should be accountable 08:17:53 sustainability 08:18:06 metrics should take energy demand into account 08:18:24 what is the impact carbon footprint use and training 08:19:56 CI in health 08:20:36 areas of interest and approach to the human way of thinking 08:20:50 BIp4Cast project 08:21:04 information from doctors, sensors devices,.. 08:21:20 Collect info about a person on their mental health 08:21:40 indicators of depression , mania,... 08:21:56 modelling mental disorders 08:22:01 Victoria Lopez' presentation: https://github.com/w3c/tpac2023-breakouts/files/12591446/Victoria.Lopez.Computational.Intelligence.Health.pdf 08:22:50 all the sensors we are using to get informartion help define a new treatment 08:23:33 updates between devices but difficulty to have data in communication, 08:24:25 solution raw data standardization 08:24:29 DiogoCortiz_ has joined #computationalintelligence 08:24:52 can data analysis predict a crisis ie bi polar disorder 08:25:05 problem of interoperability of devices 08:25:22 q+ to ask about trust 08:26:00 q? 08:26:09 privacy question for devices 08:26:30 anonymisation is done previously with patient data 08:26:47 take up item 7 08:27:05 bi polar disorder: the doctor receives an alarm 08:27:39 based on euthymic state chart 08:27:55 when is the line going up ? 08:28:08 Doctor is alerted for consultation 08:28:24 prevention is key 08:28:59 q? 08:29:51 computational intelligence as a service? How can I trust the results I get? 08:30:31 CI is to provide a means to get info quickly 08:31:16 CI models are testing the efficiency 08:32:09 it is a question of testing, posing the problem correctly 08:32:50 unpredictability of models however form of intelligence (human as well) 08:34:20 zakim, end meeting 08:34:20 As of this point the attendees have been JohnRochford, Zoltan_Kis, Maud, RachelY 08:34:22 RRSAgent, please draft minutes 08:34:23 I have made the request to generate https://www.w3.org/2023/09/13-computationalintelligence-minutes.html Zakim 08:34:30 I am happy to have been of service, RachelY; please remember to excuse RRSAgent. Goodbye 08:34:30 Zakim has left #computationalintelligence 09:01:00 tidoust has joined #computationalintelligence 09:06:09 dsinger has joined #computationalintelligence 09:06:59 dsinger has left #computationalintelligence 12:05:13 tidoust has joined #computationalintelligence