There's no connection, XOR was a representability issue. It's easy to fit a curve to a XOR dataset. That said, doing so is tremendously inefficient compared to program search over a DSL that just has NOR, for instance. So if you want a parallel, it's there.
RESEARCH
-
OpenAI Sora Team Members Are UC Berkeley BAIR Lab Graduates
By
–
The @OpenAI core team behind #Sora, Tim Brooks (
@_tim_brooks
) and Bill Peebles (
@billpeeb
), are recent graduates of @UCBerkeley @berkeley_ai BAIR Lab. Apparently they chose the name Sora because it "evokes the idea of limitless creative potential". -
Gemini’s Impressive Information Filtering from Million Token Context
By
–
Strange, imo most tester were impressed by the capability of Gemini to filter information from the ~1mio token
-
Blood Test Creates Alzheimer’s Diagnosis in Healthy People
By
–
How a blood test (p-Tau217) and drugs that have little efficacy (and major side effects) are creating "#Alzheimers Stage 1" in healthy people with no symptoms or memory deficit https://
latimes.com/science/story/
2024-02-14/inside-controversial-plan-to-diagnose-alzheimers-in-people-without-symptoms
… @MelodyPetersen @latimes @alzassociation -
Survey of Major LLM Families: GPT, Llama, PaLM
By
–
9/ Survey of LLMs – reviews three popular families of LLMs (GPT, Llama, PaLM), their characteristics, contributions, and limitations; includes a summary of capabilities and techniques developed to build and augment LLM.
-
ChemLLM: Specialized LLM Outperforms GPT-4 Chemistry Tasks
By
–
8/ ChemLLM – a dedicated LLM trained for chemistry-related tasks; claims to outperform GPT-3.5 on principal tasks such as name conversion, molecular caption, and reaction prediction; it also surpasses GPT-4 on two of these tasks.
-
V-JEPA: Self-Supervised Vision Model Trained on Two Million Videos
By
–
3/ V-JEPA – vision models trained on a feature prediction objective using 2 million videos; relies on self-supervised learning and doesn’t use pretrained image encoders, text, negative examples, reconstruction, or other supervision sources.
-
Top ML Papers Week: Sora, Gemini 1.5, Agents
By
–
The Top ML Papers of the Week (Feb 12 – Feb 18): – Sora
– Gemini 1.5
– OS-Copilot
– TestGen-LLM
– Large World Model
– LLM Agents can Hack
… -
Free MIT Linear Algebra Course for Machine Learning Foundations
By
–
Arguably, the best Linear Algebra course out there for Machine Learning! Taught by MIT's Professor Gilbert Strang. And it's FREE! Build a strong foundation in mathematics for machine learning: https://
youtube.com/playlist?list=
PL221E2BBF13BECF6C
… -
OMNI: Open-endedness Models Human Notions Interestingness
By
–
OMNI: Open-endedness via Models of human Notions of Interestingness Zhang et al.: https://
arxiv.org/abs/2306.01711 #ArtificialIntelligence #DeepLearning #MachineLearning