#MachineLearning
HIRING: Business Analyst (Machine Learning & AI) / London, England, United Kingdom https://ai-jobs.net/J45911/ #AI #MachineLearning #DataJobs #MLjobs #bigdata #DataScience #AIjobs #AIcareers #hiringnow #London #England #UnitedKingdom #Consulting #DataAnalytics #Economics #GCP
HIRING: 2023 Fall Applied Science Internship - Natural Language Processing and Speech Technologies - United States / Seattle, Washington, USA https://ai-jobs.net/J45959/ #AI #MachineLearning #DataJobs #MLjobs #bigdata #DataScience #AIjobs #AIcareers #hiringnow #Seattle #USA #ASR
HIRING: AI Developer /Remote in Ukraine https://ai-jobs.net/J44495/ #AI #MachineLearning #DataJobs #MLjobs #bigdata #DataScience #AIjobs #AIcareers #hiringnow #Ukraine #Caffe #ComputerScience #ComputerVision #DeepLearning #Keras #MachineLearning #remotedev #PyTorch #R #Theano
Happy to share the early online version of our new study (#OpenAccess): https://iopscience.iop.org/article/10.1088/1748-9326/acc81a
We use #machinelearning to detect the emergence of distinguishable climate patterns in a new model experiment simulating stratospheric aerosol injection relative to a moderate future emissions scenario.
I have a summary of the paper at https://zacklabe.com/climate-signals-and-explainable-ai/. More on this all soon! #OpenScience #SciComm
Cluster member Byung Lee is based at the University of #Vermont. He's worked in the field of #MachineLearning and #ArtificialIntelligence for decades. He shared insights with us in this #ClusterChat
Cerebras-GPT vs. LLaMA AI Model Performance Comparison
https://www.lunasec.io/docs/blog/cerebras-gpt-vs-llama-ai-model-comparison/
Discussions: https://discu.eu/q/https://www.lunasec.io/docs/blog/cerebras-gpt-vs-llama-ai-model-comparison/
Today’s AI News & Comment
Google Partners with AI Startup Replit to Take on Microsoft’s GitHub
https://www.bloomberg.com/news/articles/2023-03-28/google-partners-with-ai-startup-replit-to-take-on-microsoft-s-github
Generative AI set to affect 300 million jobs across major economies
https://arstechnica.com/information-technology/2023/03/generative-ai-set-to-affect-300-million-jobs-across-major-economies/
Microsoft plans major platform upgrades for “Windows 12” that will modernize the OS with AI, faster updates, and better security
https://www.windowscentral.com/software-apps/windows-11/microsoft-windows-corepc-modern-platform-hudson-valley-2024
1,100+ notable signatories just signed an open letter asking “all AI labs to immediately pause for at least 6 months”
https://techcrunch.com/2023/03/28/1100-notable-signatories-just-signed-an-open-letter-asking-all-ai-labs-to-immediately-pause-for-at-least-6-months/
Covariant’s CEO on building AI that helps robots learn
https://techcrunch.com/2023/03/28/covariants-ceo-on-building-ai-that-helps-robots-learn/
Everything new for Bing Chat: DALL-E images, knowledge cards, AI-generated stories, and more
https://www.msn.com/en-gb/money/-technology/everything-new-for-bing-chat-dalle-images-knowledge-cards-aigenerated-stories-and-more/ar-AA198gQI
Microsoft Bing Chat is testing 200 daily turns and 20 in-session turns
https://www.neowin.net/news/microsoft-bing-chat-is-testing-200-daily-turns-and-20-in-session-turns/?ssp=1&darkschemeovr=0&setlang=en-GB&safesearch=moderate
Will ChatGPT take your job — and millions of others?
https://www.aljazeera.com/features/2023/3/28/will-chatgpt-take-your-job-and-millions-of-others
AI changes the software-making game
https://www.axios.com/2023/03/27/ai-chatgpt-software-programming-plugins
UK to avoid fixed rules for AI – in favor of ‘context-specific guidance’
https://techcrunch.com/2023/03/29/uk-ai-white-paper/
ChatGPT comes to iPhone with Perplexity, offering its own AI assistant on iOS
https://www.imore.com/apps/iphone-apps/chatgpt-comes-to-iphone-with-perplexity-offering-its-own-ai-assistant-on-ios
#AI #ArtificialIntelligence #ML #MachineLearning #LLM #ChatGPT #BingChat
Here's a chance to use #MachineLearning for something useful. Kaggle is running a competition to develop an automatic solution that will extract data represented by four types of charts commonly found in STEM textbooks to help students with a learning, physical, or visual disability.
And you can win a portion of the $50,000 total in awards.
https://www.kaggle.com/competitions/benetech-making-graphs-accessible
#data #DataScience #DataVisualization #accessibility #a11y #kaggle #competition #hackathon #learning #stem
Fearing “loss of control,” AI critics call for 6-month pause in AI development - Enlarge / An AI-generated image of a globe that has stopped spinning. (... - https://arstechnica.com/?p=1927560 #futueoflifeinstitute #machinelearning #emilybender #elonmusk #biz #openai #gpt-4 #ai
Ars Technica: Fearing “loss of control,” AI critics call for 6-month pause in AI development https://arstechnica.com/?p=1927560 #Tech #arstechnica #IT #Technology #FutueofLifeInstitute #machinelearning #EmilyBender #ElonMusk #Biz&IT #openai #GPT-4 #AI
I tried to use Google Bard to help me with Wordle - but it didn't go well
'Inspired by Science Alert’s game of Wordle with ChatGPT I decided to put Bard to the test and play a game or two with it to see how it matches up'
#AI #ArtificialIntelligence #ML #MachineLearning #LLM #Bard #Google #WordGames #Wordle
[R] You Only Segment Once: Towards Real-Time Panoptic Segmentation [CVPR 2023]
https://github.com/hujiecpp/YOSO
Discussions: https://discu.eu/q/https://github.com/hujiecpp/YOSO
🧧 RSVP for tonight's #PrepWithGDG!
🗓️ 29.03, 10.04 - 15.05
🔗 https://gdg.community.dev/events/details/google-gdg-cloud-munich-presents-certification-study-group-virtual-info-session/
With Fady https://www.linkedin.com/in/fadyibra/ as guest speaker!
Is a series of virtual event dedicated to certification in Professional #MachineLearning Engineer:
Join us on 𝐌𝐨𝐧𝐝𝐚𝐲, 𝐀𝐩𝐫𝐢𝐥 3, at 4 𝐏𝐌 𝐂𝐄𝐒𝐓, for our webinar explaining the new Call for Application launched by AION Labs and @BioMed_X - it's all about applying #AI to enable #Molecular Glue #Therapies! 💻💊
Register here: https://us02web.zoom.us/webinar/register/WN_z2Aj4K5GRgGR7YVaKmoeew

Preprint of "Describing and Organizing #SemanticWeb and #MachineLearning Systems in the SWeMLS-KG" (to be presented @eswc_conf) is available at https://arxiv.org/abs/2303.15113 with @fajarjuang@twitter.com, M. Llugiqi, @MartaSabou@twitter.com, A. Ekelhart,
@heikopaulheim@twitter.com, A. Breit,
@revenkoartem@twitter.com,
@LaWaltersdorfer@twitter.com, @reddines@twitter.com
The corresponding
@orkg observatory with 475 comprehensively described papers can be found here: https://orkg.org/observatory/Neurosymbolic_artificial_intelligence
[D] Do model weights have the same license as the modem architecture?
https://github.com/Lightning-AI/lit-llama
Discussions: https://discu.eu/q/https://github.com/Lightning-AI/lit-llama
POMA
This study used #MachineLearning to better understand the #soundscapes in daycare centers. https://doi.org/10.1121/2.0001708
Last day at #AGBTAg! Don’t miss the talk by Rupashree Dass: “Machine learning-based phenotype prediction help in detecting significantly associated genetic markers” in the Independence Ballroom at 4:25 pm! #machinelearning #AGBTAg23 #predictivebreeding #measurablybetter

date: 2023-03-29 13:16:36
by: AICareer (🚀Jobs-Internships-Scholarships)
PhD in Operations Research and Machine Learning in Healthcare
at University of Amsterdam (UvA)
Check the details here: https://ai-jobs.org/job/phd-in-operations-research-and-machine-learning-in-healthcare/
#phd #phdposition #phdscholarship #operationsresearch #MachineLearning #university #universityofamsterdam #ai #artificialintelliegence
🐦🔗: https://twitter.com/twitter/statuses/1641066830293790720
#PhdPosition

I found this one particularly interesting based on the hardware in the background. Is this what #MachineLearning across millions of #HigherEd / #EdTech images reveals; 1970s-era technology?

RT toncarrera: One of the best qualities of the human being lies in its #desire to find #answers, without it, it is very possible that we would not have advanced as we do.
#datascience #MachineLearning #BigData #DeepLearning #ArtificialIntelligence #dataengineering #datascientist #innovation

date: 2023-03-29 11:01:46
by: SIBBM
Are you passionate about #MolecularMedicine and #Bioinformatics? 🔬💻 The University of Turin is offering a #PhDPosition in these fields, with a particular emphasis on applying #MachineLearning techniques to biological source data.
Learn more at https://www.sibbm.org/job-offers 🌟
🐦🔗: https://twitter.com/twitter/statuses/1641032899557220352
#PhdPosition
.
.
.
.
.
.
#artificialintelligence #aiart #generativeart #digitalart #neuralnetwork #deepdream #deepart #machinelearning #AI #inspiration #scifiart #scifi #randomart #random

🚀Kornia 0.6.11 is out!
🔥Amazing DISK local feature by Michał Tyszkiewicz
(used in #IMC2021-winning solutions)
🔥 Random MedianBlur, RandomSnow, RandomRain augs
✅many 🐞-fixes
📚Release notes:
https://github.com/kornia/kornia/releases/tag/v0.6.11
#computervision #machinelearning #ai #deeplearning #PyTorch #opensource

We are hiring! If you’re interested in doing a PhD in machine learning and information field theory, apply here:
If you're at #QConLondon today, hang around to the end. I'll be closing out the conference today talking about "Malignant Intelligence," prompt engineering and software archeology. See you there? https://qconlondon.com/keynote/mar2023/malignant-intelligence #MachineLearning #GenerativeAI #AI

Building a better mousetrap.
Berkeley Earth's historical temperature analysis is moving into a new era with a new high-resolution reconstruction, available now as a beta product.
More information available at the blog post: https://berkeleyearth.org/introducing-the-berkeley-earth-high-resolution-dataset/
#GlobalWarming #ClimateChange #Temperature #MachineLearning #Climate

If you are performing machine learning in areas such as finance or science, it can sometimes be disconcerting for management to see that the output data varies every time the learning is run with the same input data. This doesn’t instil any confidence that your solution is working correctly. Here are some tips to make results deterministic.
https://mrprediction.com/making-machine-learning-deterministic/
[N] OpenAI Vendor Lock-in: The Ironic Story of How OpenAI Went from Open Source to "Open Your Wallet", and some FOSS alternatives you can try right now.
https://www.lunasec.io/docs/blog/openai-not-so-open/
Discussions: https://discu.eu/q/https://www.lunasec.io/docs/blog/openai-not-so-open/
HIRING: Data Scientist - Intern / Canada https://ai-jobs.net/J45798/ #AI #MachineLearning #DataJobs #AIcareers #MLjobs #bigdata #DataScience #AIjobs #hiringnow #Canada #Bayesian #Causalinference #ComputerScience #Ecommerce #Econometrics #Engineering #Genetics #MachineLearning
HIRING: Director of Analytics Engineering / United States https://ai-jobs.net/J45786/ #AI #MachineLearning #DataJobs #AIcareers #MLjobs #bigdata #DataScience #AIjobs #hiringnow #UnitedStates #APIs #Architecture #CICD #Engineering #Python #Snowflake #SQL #Unstructureddata
HIRING: Data Scientist, Finance / Remote (Global) https://ai-jobs.net/J45803/ #AI #MachineLearning #DataJobs #AIcareers #MLjobs #bigdata #DataScience #AIjobs #hiringnow #RemoteJob #Global #Blockchain #ComputerScience #Crypto #Dataanalysis #Datamanagement #Datapipelines #Economics
I've been writing some blog posts aroud AI and the training algorithms. Phew. It takes quite a bit of time to really understand these concepts. A lot of them take a lot of concepts from statistics. I was never really good at statistics. The math doesn't really excite me. I mean lambda calculus is pretty hot.
The thing is - you'll need a solid background in math and statistiics to really "get" AI and training models. It still feels like black magic to me. #AI #machinelearning #ML
AI In The Sky Closing In On Pest Plants [New Zealand]
--
https://www.farmersweekly.co.nz/technology/ai-in-the-sky-closing-in-on-pest-plants/ <-- shared article
--
#GIS #spatial #mapping #machinelearning #business #ai #management #environment #remotesensing #imagery #spatialanalysis #NewZealand #invasivespecies #invasiveplants #invasive #pests #pestcontrol #artificialintelligence #productivity #croplands #agriculture #farming #model #modeling #biosecurity #threats #Nassella #tussock #spatiotemporal #appliedscience
Lynker Analytics | Environment Canterbury




[P] 🎉 Announcing Auto-Analyst: An open-source AI tool for data analytics! 🎉
https://github.com/aadityaubhat/auto-analyst
Discussions: https://discu.eu/q/https://github.com/aadityaubhat/auto-analyst
Mark your calendars for our April meeting! Serving #MachineLearning Models at Scale with #FastAPI.
Compare and contrast #MachineLearning visualizations of multiple models' "attention" using Kangas and its new metric mask.
See the live demo: https://kangas.comet.com/?datagrid=/data/attention.datagrid&select=ID,Image,Model+Name
Try "Group by" ID to compare images; try "Group by" Model Name to compare models.
Open source from the folks at https://comet.com/

🤔 Prediction: following #microsoft security copilot announcement, every #security vendor will be launching gpt4 integration by end of 2023. We are in the era of ML infused security https://www.bleepingcomputer.com/news/microsoft/microsoft-brings-gpt-4-powered-security-copilot-to-incident-response/ #infosec #machinelearning
Game changer 👍🏻 Microsoft today announced #Security Copilot, a new #ChatGPT-like assistant powered by AI that takes advantage of Microsoft's threat intelligence footprint to make faster decisions during incident response and to help with threat hunting and security reporting. Copilot answers defenders' security questions via a ChatGPT-like interface and continuously learns to adapt to each environment to advise on the best course of action. https://www.bleepingcomputer.com/news/microsoft/microsoft-brings-gpt-4-powered-security-copilot-to-incident-response/ #infosec #machinelearning
How does AI change programming languages?
https://blog.darklang.com/gpt/
Discussions: https://discu.eu/q/https://blog.darklang.com/gpt/
Released 👍🏻 Cerebras open sources seven GPT-3 models from 111 million to 13 billion parameters. these models set new benchmarks for accuracy and compute efficiency. Today’s release is designed to be used by and reproducible by anyone. All models, weights, and checkpoints are available on Hugging Face and GitHub under the Apache 2.0 license. The Cerebras CS-2 systems used for training are also available on-demand via Cerebras Model Studio https://www.cerebras.net/blog/cerebras-gpt-a-family-of-open-compute-efficient-large-language-models/ #machinelearning
Today’s AI News & Comment
Google's claims of super-human AI chip layout back under the microscope
https://www.theregister.com/2023/03/27/google_ai_chip_paper_nature/
Meet Dolly - A new open-source ChatGPT clone
https://www.techradar.com/news/meet-dolly-a-new-open-source-chatgpt-clone
Romance Scammers are Using AI Images to Trick Victims
https://petapixel.com/2023/03/28/romance-scammers-are-using-ai-images-to-trick-victims/
Clearview AI used nearly 1m times by US police, it tells the BBC
https://www.bbc.co.uk/news/technology-65057011
How a Senator’s Misguided Tweet Can Help Us Understand AI
https://gizmodo.com/senator-chris-murphy-comments-on-ai-chatgpt-regulation-1850270985
Photo Editing App Bazaart Adds New AI Background Feature
https://petapixel.com/2023/03/28/photo-editing-app-bazaart-adds-new-ai-background-feature/
Microsoft lets generative AI loose on cybersecurity
https://techcrunch.com/2023/03/28/microsoft-lets-generative-ai-loose-on-cybersecurity/
The US government is gearing up for an AI antitrust fight
https://www.theverge.com/2023/3/28/23660101/ai-competition-ftc-doj-lina-khan-jonathan-kanter-antitrust-summit
Newegg’s AI bot tried to help me build a PC - and it didn’t go well
https://www.techradar.com/news/neweggs-ai-bot-tried-to-help-me-build-a-pc-and-it-didnt-go-well
ChatGPT is being used to lure victims into downloading malware
https://www.techradar.com/news/chatgpt-is-being-used-to-lure-victims-into-downloading-malware
#AI #ArtificialIntelligence #ML #MachineLearning #LLM #ChatGPT #BingChat
Couch Potato (or, as its end users knew it, Perspective) was a flagship #MachineLearning tool built by the Biden engineering team.¹ Political campaigns are under-staffed, under-funded, and frankly, under duress. AI systems can and do replace the work otherwise distributed to an army of employees.²
¹ https://www.nytimes.com/2023/03/28/us/politics/artificial-intelligence-2024-campaigns.html
² https://mastodon.social/@MattHodges/109761554331474164

Make Use Of: Do ChatGPT Responses Have a Character or Word Limit? https://www.makeuseof.com/do-chatgpt-responses-have-character-or-word-limit/ #Tech #MakeUseOf #TechNews #IT via @morganeogerbc #ArtificialIntelligence #TechnologyExplained #MachineLearning #ChatGPT #Chatbot
I work as an MLE doing time-series classification. The Perceiver IO model has been a silver bullet for performance - so I made a code walkthrough [P]
Discussions: https://discu.eu/q/https://youtu.be/Xe7VT8-kDzg
Interesting solution 👉 Automate Amazon Rekognition Custom Labels model training and deployment using AWS Step Functions https://aws.amazon.com/blogs/machine-learning/automate-amazon-rekognition-custom-labels-model-training-and-deployment-using-aws-step-functions/ #AWS #MachineLearning #DevOps #MLOps

What are your feelings towards #MachineLearning these days?
AI-enhanced development makes me more ambitious with my projects
https://simonwillison.net/2023/Mar/27/ai-enhanced-development/
Discussions: https://discu.eu/q/https://simonwillison.net/2023/Mar/27/ai-enhanced-development/
Tech charlatans and U.S. Senators are now spreading misinformation about predictive AI tools, which are not sentient.
https://www.vice.com/en/article/ak3zbp/for-the-love-of-god-ai-chatbots-cant-decide-to-do-anything
#ChatGPT #chatbots #AI #ArtificialIntelligence #Disinformation #machinelearning #largelanguagemodels #OpenAI
[N] OpenAI may have benchmarked GPT-4’s coding ability on it’s own training data
https://aisnakeoil.substack.com/p/gpt-4-and-professional-benchmarks
Discussions: https://discu.eu/q/https://aisnakeoil.substack.com/p/gpt-4-and-professional-benchmarks
[P] two copies of gpt-3.5 (one playing as the oracle, and another as the guesser) performs poorly on the game of 20 Questions (68/1823).
https://evanthebouncy.medium.com/llm-self-play-on-20-questions-dee7a8c63377
Discussions: https://discu.eu/q/https://evanthebouncy.medium.com/llm-self-play-on-20-questions-dee7a8c63377
I've been doing some ML work lately, and in my darkest moments, YouTube Music tosses Van Morrison into my De La Soul radio, and I remember nobody knows what they're doing.
#Zoom announces #openai integration for smart meetings, closing the gap with #microsoft teams and putting #google meet, #webex and blue jeans on notice 🤔 Evolving Zoom IQ, our smart companion, with new features and a collaboration with OpenAI https://blog.zoom.us/zoom-iq-smart-companion/ #machinelearning #productivity
WEBINAR THURSDAY 3.30
Threat Modeling and Machine Learning. Huh?
Or When with chatGPT replace Adam Shostack?
Reserve your spot today: https://www.iriusrisk.com/impact-machine-learning

@jbhayet trabaja en #IA en el #CIMAT
@Nick_Stevens_graphics hace animaciones bien chidas de naves espaciales
y puedes buscar hashtags de tu interés como #astronomy #ai #machinelearning o que se yo y segurlos (así es, aquí puedes seguir hashtags)
#python question:
I am using opencv face detection algo and it works great but the cropped images are different sizes as the input images are different in sizes too.
Do i need to resize them before I feed them to the “haarcascade_frontalface_alt2.xml” or is there a built in parameter i can use to get the images resized?
Havent found anything yet online…
Code here:

Just update rules between neurons
https://www.jtolio.com/2023/03/just-update-rules-between-neurons/
Discussions: https://discu.eu/q/https://www.jtolio.com/2023/03/just-update-rules-between-neurons/
Todays insight on #ArtificialIntelligence from
#ChatGPT #GPT4 #AI #MachineLearning #DeepLearning #NeuralNetworks #BigData

[R] New article in Nature Medicine describes the risks of using an AI- & ML-based tool known as NarxCare to guide opioid prescription decision making
Discussions: https://discu.eu/q/https://rdcu.be/c8sw9
#machinelearning but to prove isomorphism
[D] Favorite tips for staying up to date with AI/Deep Learning research and news?
https://sebastianraschka.com/blog/2023/keeping-up-with-ai.html
Discussions: https://discu.eu/q/https://sebastianraschka.com/blog/2023/keeping-up-with-ai.html
First was an excellent talk by Eric Xing on meta-learning and "lego-learning" at #CMU. I loved the unified #MachineLearning approach discussed here, as well as the compositional strategy for #ML in real world systems. Highly recommend https://www.youtube.com/watch?v=dmjMnoDkqnU (2/7)
What's being ignored most of the time in all the excitement surrounding AI is that creation is about the joy of creating, the pleasure of developing better skills, of surpassing your former self, discovering new techniques, surprising yourself and others.
[P] New toolchain to train robust spiking NNs for mixed-signal Neuromorphic chips
https://arxiv.org/abs/2303.12167
Discussions: https://discu.eu/q/https://arxiv.org/abs/2303.12167
I can clearly imagine this person's dilemma:
https://www.reddit.com/r/blender/comments/121lhfq/i_lost_everything_that_made_me_love_my_job/
Great first reply too:
"In a society that judges a person's worth purely on their economic contribution it's going to be a disaster."
Sooner or later AI will supersede us in every area, making more and more jobs obsolete because AI is cheaper and more efficient than human labor. I.m.h.o. the best way to go is a universal basic income, so we don't need to compete with AI.
ChatGPT isn’t hallucinating; it’s delirious
https://digressionsnimpressions.typepad.com/digressionsimpressions/2023/03/the-delirium-of-llms.html
https://www.linkedin.com/posts/alighodsi_hello-dolly-democratizing-the-magic-of-chatgpt-activity-7045014694385917952-JUVH?utm_source=share&utm_medium=member_desktop "We are open sourcing Dolly, a ChatGPT-like model that can do instruction following, brainstorming, summarization. The remarkable thing is that we created it for $30 using one server for 3 hours on a small dataset using a 2 year old open source LLM model. The secret in magical human-like interactivity probably lies in a small dataset, not huge models, not training for 100,000 GPU hours." #StochasticParrots #LLM #MachineLearning
I feel like AI is something that's passing me by and I really want to learn about it. Any recommendations for starting from scratch? #ai #openai #tech #ml #machinelearning
This visualization is of the 1st convolutional layer (evolution of its weights) in a CNN trained on classification of handwritten digits.
#machinelearning #deeplearning #visualization #creativecoding

"The problem is that, in our world, artists have to make a living through their art by selling it, and so they have to think about its market value. We’re introducing a technology that can utterly wreck people’s livelihoods, and in a free-market economic system, if your skills decline in value, you’re screwed."
https://jacobin.com/2023/03/ai-artificial-intelligence-art-chatgpt-jobs-capitalism/
Today’s AI News & Comment
WGA Would Allow Artificial Intelligence in Scriptwriting, as Long as Writers Maintain Credit
https://variety.com/2023/biz/news/writers-guild-artificial-intelligence-proposal-1235560927/
OpenAI is massively expanding ChatGPT’s capabilities to let it browse the web and more
https://www.theverge.com/2023/3/23/23653591/openai-chatgpt-plugins-launch-web-browsing-third-party
ChatGPT plugins are officially here, and I’m already filled with dystopian dread
https://www.techradar.com/news/chatgpt-plugins-are-officially-here-and-im-already-filled-with-dystopian-dread
Voice Deepfakes Are Calling. Here's How to Avoid Them
https://gizmodo.com/ai-deepfake-voice-how-to-avoid-spam-phone-calls-1850245346
How to build your own ChatGPT alternative on the cheap? There's one way
https://www.techradar.com/news/how-to-build-your-own-chatgpt-alternative-on-the-cheap-theres-one-way
It wasn't just you, ChatGPT just fixed a 'significant' bug
https://www.techradar.com/news/it-wasnt-just-you-chatgpt-just-fixed-a-significant-bug
Levi’s to Use AI-Generated Models to ‘Increase Diversity’
https://petapixel.com/2023/03/24/levis-to-use-ai-generated-models-to-increase-diversity/
Microsoft’s Bing AI Chatbot is actually tempting users away from Google
https://www.techradar.com/news/microsofts-bing-ai-chatbot-may-be-tempting-users-away-from-google
#AI #ArtificialIntelligence #ML #MachineLearning #LLM #ChatGPT #BingChat
New AI research & news everywhere! A short post on my personal approach to keeping up with things.
https://sebastianraschka.com/blog/2023/keeping-up-with-ai.html
heise+ | Teachable Machine: Eigenes KI-Modell trainieren mit Tools von Google
Ein Dienst von Google macht es zum Kinderspiel, mit Machine Learning eigene KI-Modelle zu erstellen. Wir zeigen, wie das schon mit einem mittelmäßigen PC geht.
#Bilderkennung #Google #KünstlicheIntelligenz #MachineLearning #RaspberryPi #news

Yup. #MachineLearning #LLM #ChatGPT -> “Google and Microsoft’s chatbots are already citing one another in a misinformation shitshow - The Verge”
https://www.theverge.com/2023/3/22/23651564/google-microsoft-bard-bing-chatbots-misinformation
Ubisofts Ghostwriter looks like a tool where AI can really help authors, especially since game worlds are getting bigger and bigger:
#game #development #ai #machineLearning #generative #ubisoft
"Microsoft’s Bing said Google’s Bard had been shut down after it misread a story citing a tweet sourced from a joke. It’s not a good sign for the future of online misinformation."
https://www.theverge.com/2023/3/22/23651564/google-microsoft-bard-bing-chatbots-misinformation
#miscrosoft #bing #ChatGPT #bard #google #ai #MachineLearning #news #tech
MICROSOFTS NEW Insane BING Image Creator SHOCKS The Entire Industry!
Ignore the clickbait headline. It's actually a serious review.
I gave it a go in Bing Chat in the Edge Dev browser, results shown below.
#AI #ArtificialIntelligence #ML #MachineLearning #LLM #ChatGPT #BardABing
https://www.youtube.com/watch?v=7aekr13tdKI
14 minute video

"When we asked Bard to generate an image of an "orange cat," the AI did what it does best: spat out a string of text. The only problem with this output is, it seems, that the chatbot thinks it's an image."
Great article from Josh W. Comeau about if AI will replace developers:
https://www.joshwcomeau.com/blog/the-end-of-frontend-development/
Today, @mozilla announced https://mozilla.ai, a "startup — and community — building trustworthy and open-source AI".
Exciting news in the world of tech! The latest advancements in machine learning and AI are revolutionizing industries like healthcare, finance, and transportation. With data-driven insights and predictive analytics, we can make more informed decisions and drive innovation like never before. What technology are you most excited about? #AI #MachineLearning #Innovation
So this is where the images scraped from Adobe's users are going...
https://www.adobe.com/sensei/generative-ai/firefly.html
#adobe #ai #ArtificialIntelligence #ml #MachineLearning #DeepLearning
It's official! We've launched a toolkit to help teachers and students adapt classes to the sudden rise of generative AI. Browse recommended policies, watch tutorials, and learn strategies for integrating—or excluding—text- and media-generators in your classes.
https://umaine.edu/learnwithai
#EdChatme #EdChat #EdTech #STEMEducation #HigherEd #Edutooters #AcademicMastodon #OpenAI #ChatGPT #DALLE #GenerativeAI #AIimages #AIethics #GenerativeAIArt #DeepLearning #MachineLearning #llm #plagiarism
So ChatGPT is basically just natural language programming where you don’t know what data and algorithms you’re working with and can’t predict the exact result you’re going to get.
Sure sounds like progress to me!
🤓👍
#ChatGPT #openAI #artificialIntelligence #ai #machineLearning #ml #largeLanguageModels #llm #statisticalModels #probabilisticModels #glorifiedMatrixMultiplication
👋 🌱 🐝 🤖 🛡️⛅ 👁️ 📚 🔖 🎯 🎥 🎤
Love to learn together, and follow the Ops in Dev newsletter?
Subscribe at https://buttondown.email/opsindev.news
Looks like Bing reindexed my profile, can someone search for it using Bing Chat? I would like to see if the injection works, but I no longer have access to it, I'm banned lol
Today’s AI News & Comment
Microsoft’s new share button makes it easy to show people what Bing AI is saying
https://www.theverge.com/2023/3/17/23645303/microsoft-bing-ai-chat-share-button-faster
AI can fool voice recognition used to verify identity by Centrelink and Australian tax office
https://www.theguardian.com/technology/2023/mar/16/voice-system-used-to-verify-identity-by-centrelink-can-be-fooled-by-ai
ChatGPT is great — you’re just using it wrong
https://www.salon.com/2023/03/18/chatgpt-is-great--youre-just-using-it_partner/
OpenAI CEO, CTO on risks and how AI will reshape society (21'37" video from abc News)
https://www.youtube.com/watch?v=540vzMlf-54
#AI #ArtificialIntelligence #ML #MachineLearning #LLM #ChatGPT #BingChat
Will AI Replace Programmers? A Look at the Advancement of Artificial Intelligence https://magazine.mindplex.ai/mp_news/will-ai-replace-programmers-a-look-at-the-advancement-of-artificial-intelligence/ #artificialintelligence #ai #machinelearning #technology #deeplearning #programming #tech #robotics
These visualizations are made in the same way, but visualize two layers (evolution of their weights) of a CNN trained on classification of handwritten digits.
First convolutional layer (left img) and last fully connected layer (right img).
#machinelearning #deeplearning #visualization


Large, undocumented datasets are known to introduce harmful biases in AI systems. But how does the model itself spread prejudices? On Denotation IO this week: "the ethics of generalization", covering some fundamentals of machine learning, and their impact on a system’s social behavior.
A call for building small, dynamic models with strong semantic capabilities.
In 2018, I created visualizations of neural network weight evolution during training. Some of the results are pretty, so I'll share them here once in a while.
This visualization is of Word2Vec's 50th neuron (evolution of its weights).
#machinelearning #deeplearning #visualization

heise+ | KI-Kunst ausprobiert: Mit Midjourney eindrucksvolle Werke schaffen
Midjourney erzeugt beeindruckende Bilder. Spielend imitiert das KI-Tool Künstler und setzt fast jede Idee um. Wie das konkret geht, erklärt unser Ratgeber.
Personal observation: in the next five years companies will either become machine learning companies, or will be disrupted by a machine learning company #machinelearning #business