What is my math transformer doing? — Three results on interpretability and generalization Charton et al.: https://
arxiv.org/abs/2211.00170 #MachineLearning #DeepLearning #ArtificialIntelligence
RESEARCH
-
Math Transformer Interpretability and Generalization Research
By
–
-
AI Predicts Skin Cancer Recurrence with Machine Learning
By
–
Artificial Intelligence Method Can Help Predict Skin Cancer Recurrence https://
healthitanalytics.com/news/artificia
l-intelligence-method-can-help-predict-skin-cancer-recurrence
… @HITAnalytics #Healthcare #AI #DataScience #Python #Robots #BigData #Analytics #100DaysofCode #IoT #serverless #womenwhocode #DeepLearning #DigitalTransformation #MachineLearning -
MIT CSAIL Talk Summary Shared by Researcher
By
–
Oh wow, this is a very nice summary thread of my talk today @MIT_CSAIL . Thank you Nur😍 https://t.co/77HtuimQ27
— Fei-Fei Li (@drfeifei) 3 novembre 2022Oh wow, this is a very nice summary thread of my talk today @MIT_CSAIL . Thank you Nur
-
AI Generated Academic Papers: Biology, Economics, Physics, ML
By
–
What AI thinks we academics publish… I asked MidJourney to generate a biology, economics, physics, and a machine learning paper. They look like they are from an alternate universe where Reviewer 3 is an Eldritch Horror (aren’t they all), but I bet you can guess which is which.
-
Blindsight: The AI Future We Should Avoid
By
–
(Of all of the science fiction books we could be living in, Blindsight would not be my novel of choice.) Paper here:
-
Measuring Willingness to Pay: Methods for Pricing Strategy
By
–
I assume this is mostly being done for fun (right?) but entrepreneurs & marketers should know that you cannot assess willingness to pay with a poll. This paper gives a menu of methods to figure out pricing, plus a useful chart of pros & cons for each. https://
researchgate.net/publication/24
2382759_A_Review_of_Methods_for_Measuring_Willingness-to-Pay
… -
Adversarial Policy Defeats KataGo Go AI System 99%
By
–
Befuddling AI Go Systems: MIT, UC Berkeley & FAR AI’s Adversarial Policy Achieves a >99% Win Rate Against KataGo https://
syncedreview.com/2022/11/02/bef
uddling-ai-go-systems-mit-uc-berkeley-far-ais-adversarial-policy-achieves-a-99-win-rate-against-katago/
… -
Iterative approach to building and refining positive datasets
By
–
from my own experience you want something interactive and change your mind around quite a bit. so you're building the positive set, seeing the results, then tweaking your positive set over time. it's an incremental iterative thing.
-
Raw Scalable Intelligence: The Next AI Frontier
By
–
Pretty pictures are cool. But raw, scalable intelligence is on its way.
-
Beyond GPT-3: Next Wave of Superhuman AI Systems
By
–
In a couple of years, we’re going to look back at GPT-3 as a fun toy… The coming waves of AI systems will be smarter than humans in many ways. A different kind of intelligence. It’ll feel alien, yet familiar. And more generally capable than any single person alive today.