Congrats to @UWCheritonCS colleagues Niki Hasrati (
@niki_hasrati
) & Shai Ben-David (
@shaibendavid5
) on best paper at #ALT2023. Main result shows when an online learner must be computable, it's no longer characterized by Littlestone dimension. Check it out: https://
arxiv.org/abs/2302.04357
RESEARCH
-
Computability Constraints Redefine Online Learning Theory Characterization
By
–
-
Theory of Mind Experiments on LLMs Show Promising Results
By
–
These theory of mind experiments on LLMs are surprising and quite fascinating (also BLOOM is pretty strong) https://
t.co/iZ2llu4vxm -
ChatGPT and Bing’s Arithmetic Capabilities: Marcus Counterexamples
By
–
Contraejemplos de Gary Marcus, parte 1. Contar y problemas aritméticos sencillos bien. Incluso algunos de los fallos anteriores de ChatGPT ya no ocurren. Pero eso sí, las respuestas de Bing son otro nivel, con emojies incluidos.
-
GPT’s remarkable ability to generalize with invented words
By
–
Aún recuerdo del paper original de GPT-3 lo impresionante que me resultaban los experimentos donde el modelo entendía y operaba con palabras inventadas, demostrando así su capacidad de generalizar su conocimiento. Ahora ya estamos aquí, y no deja de ser igualmente impresionante!
-
Aspect-based Sentiment Analysis: Modeling Approaches and Techniques
By
–
Aspect-based sentiment is stuff like “service: good, food: bad”. Sometimes you can model it as multiple label vectors on the text, other times it’s more like spancat or relation extraction
-
Data Engineering vs Data Science: Infrastructure and Analysis
By
–
Data Engineering deals with the practical aspects of collecting, storing, & processing large amounts of data, while Data Science involves using that data to build models and make data-driven decisions. Simply put, DE focuses on the infrastructure while DS focuses on the analysis.
-
Data Engineering: The Essential Foundation for Data Science
By
–
All paths to Data Science leads through Data Engineering.
— Shubham Saboo (@Saboo_Shubham_) 11 février 2023
Credits: @christianbdata pic.twitter.com/BgtB2T3SAIAll paths to Data Science leads through Data Engineering. Credits: @christianbdata
-
AI Deployment and Gender Equality Gap in STEM
By
–
Will the deployment of AI narrow the gender equality gap or widen it? This International Day Of Women And Girls In Science, let’s critically analyze the current state of women's participation in STEM and bridge the gap. @yaasnadua @jyotij0shi @McKinsey @Walmarttech @kroop_ai
-
Women in AI: Gender Gap in Science and Future Labour Market
By
–
According to a report by @UNESCO
, only 29% of global positions in science R&D are occupied by women. Hence the question arises if technologies such as AI continues to mature, how will the future labour market look for women? -
Skepticism about prompt vs human-written policy in ChatGPT RLHF
By
–
Thought about this more: It doesn’t make sense it’s the prompt, because inference costs. Still feels human-written — but why would you RLHF into reciting all policy verbatim? ChatGPT falls for the same trick and the text it recites is just parameters.