Continuous Learning_Startup & Investment – Telegram
Continuous Learning_Startup & Investment
2.4K subscribers
513 photos
5 videos
16 files
2.72K links
We journey together through the captivating realms of entrepreneurship, investment, life, and technology. This is my chronicle of exploration, where I capture and share the lessons that shape our world. Join us and let's never stop learning!
Download Telegram
RIP my legend and teacher, Charlie Munger.

Thank you so much.
"I think a life properly lived is just learn, learn, learn all the time."

— Charlie Munger
딥마인드가 단백질 구조예측, 기상 예측에 이어서 물질 구조예측을 내놨는데 이것도 그래프 네트워크를 활용했네요.

Today at Google DeepMind we’re introducing Graph Networks for Materials Exploration (GNoME) - a computational tool that has already led to the discovery of 2.2 million new crystals. Crystals are essential materials to our everyday lives, from solar cells to computer chips.

Given that roughly 28,000 stable materials were discovered in the last decade, this new discovery is equivalent to nearly 800 years of knowledge!

#GNoME was trained using a technique called ‘active learning’ and helps to predict the stability of the materials, and therefore pinpoint the most usable ones for real-world applications. The work also represents a leap forward in efficiency for this type of model - this could have a significant impact on how much compute is required to discover a new material.

We believe scaling up materials discovery breakthroughs like this could help us tackle some of the major challenges of the 21st century - like finding cleaner forms of producing electricity.  Read in our blog post about what this research could mean for the future of materials science:

https://deepmind.google/discover/blog/millions-of-new-materials-discovered-with-deep-learning/

GNoME shows the potential of using AI to discover and develop new materials at scale. External researchers in labs around the world have independently created 736 of these new structures experimentally in concurrent work.
OpenAI 코파운더인 Greg Brockman이 트윗으로 올렸던 내용.
"놀랍게도 GPT-4에서 생성된 영상의학 보고서 요약이 어떤 경우에는 숙련된 영상의학 전문의가 작성한 보고서 요약보다 선호된다는 것을 발견했다."
어쩌면 약간 침소봉대하는 내용일수도 있는데, 어쨌든 MS와 함께 EMNLP 2023에 발표했던 논문에서 그렇다고 하네요. arXiv에는 10월말에 공개했던 내용이기도.... 아마 RSNA에 가신 분들이 많아서 조용히 넘어가는 듯 싶네요.
제목:
Exploring the Boundaries of GPT-4 in Radiology
요약:
최근 일반 도메인 대규모 언어 모델(LLM)의 성공으로 자연어 처리 패러다임이 여러 도메인과 애플리케이션에 걸쳐 통합된 기반 모델로 크게 변화했습니다. 이 백서에서는 지금까지 가장 성능이 뛰어난 LLM인 GPT-4의 성능을 영상의학 보고서의 텍스트 기반 애플리케이션에서 평가하여 최첨단(SOTA) 영상의학 전용 모델과 비교하는 데 중점을 둡니다. 다양한 프롬프트 전략을 탐색하면서 다양한 범위의 일반적인 방사선 작업에서 GPT-4를 평가한 결과, GPT-4가 최신 SOTA 방사선 모델보다 성능이 뛰어나거나 동등한 수준인 것으로 나타났습니다. 제로 샷 프롬프트를 통해 GPT-4는 이미 시간적 문장 유사성 분류(정확도) 및 자연어 추론(F1)에서 방사선학 모델보다 상당한 이득(≈ 10% 절대적 개선)을 얻었습니다. 데이터 세트별 스타일이나 스키마를 학습해야 하는 작업(예: 결과 요약)의 경우, GPT-4는 예제 기반 프롬프트를 통해 개선되고 감독된 SOTA와 일치합니다. 보드 인증 방사선 전문의와 함께 실시한 광범위한 오류 분석에 따르면 GPT-4는 미묘한 도메인 지식이 필요한 복잡한 맥락에서 가끔 발생하는 오류만 제외하고 충분한 수준의 방사선학 지식을 갖추고 있는 것으로 나타났습니다. 결과를 요약하면, GPT-4의 결과물은 기존의 수기로 작성된 인상과 전반적으로 비슷한 수준인 것으로 나타났습니다.
https://www.microsoft.com/en-us/research/blog/gpt-4s-potential-in-shaping-the-future-of-radiology/?fbclid=IwAR0gfYDT9lX4YSjMhnaYU17DqAvQELAM7HwrS1prf4amwBi1PEEs_ZNIqD0
27살에 Varda라는 우주 기업을 공동 창업하고 & 동시에 파운더스 펀드에서 파트너로 일하고 있는 Delian Asparouhov가 30살을 마주하면서 남긴 글.

https://delian.io/thirty
👍1
그저께 PIKA 1.0이 나왔는데

오늘은 오픈소스 Video Generation 오픈소스인 Vchitect 라는 프로젝트가공개되었네요. T2V, I2V 두 모델을 공개했습니다.

T2V
- 코드: https://t.co/sT24IvlhD2
- 스페이스: https://t.co/bhF6TDUZ67

I2V
- 코드: https://t.co/IryTCXbiiJ
- 스페이스: https://t.co/joJ2jUMmZB

Video Generation도 이제 치열한 경쟁이 시작되겠군요. 과연.. 현타가 올 기업들이 얼마나 생길지..
Khosla got a full scholarship to Carnegie-Mellon University in Pittsburgh for a master’s in biomedical engineering, graduating in 1978. After that, he wanted to go to Stanford Business School but was rejected. He applied again, only to be waitlisted. In a pattern that would be repeated many times, he didn’t take maybe for an answer. He called the admissions office repeatedly, finally just driving across the country and showing up. The details are complicated, but he got in.

One historic anecdote from his Kleiner days concerns early search engine and web portal Excite, which Khosla had invested in for the firm. In early 1999, “Vinod tried to get the founders of Excite to buy Google for $2 million,” said Doerr, who believes Google co-founders Larry Page and Sergei Brin might have accepted the offer. “But they would only offer Google $1 million. It drove Vinod crazy. But it then left the opportunity for Kleiner to go ahead and invest in Google.”

“he’s the person who can see the future, see around corners. That’s his magic weapon. He can see what other people don’t.”
Continuous Learning_Startup & Investment
RIP my legend and teacher, Charlie Munger. Thank you so much.
Remembering my teacher Charlie Munger

Thursday, November 30, 2023

I was on a business trip in Asia on Tuesday when I got the call from the Munger family informing me that Charlie was in his final hours. I hopped onto the next flight I could find to California, and before departure, was able to talk to Charlie through the help of his daughter. Charlie had largely lost consciousness, but still I could clearly hear him trying to make a sound to acknowledge he had heard me. Upon landing, I learned that Charlie had left us a few hours earlier.

I arrived at his Santa Barbara home and had the opportunity to spend cherished time with family members, reminiscing about all things Charlie. Charlie was engaging, humorous and full of wit even at Thanksgiving dinner just a few days ago, family members told me. I visited his home library again. In that very room, exactly 20 years ago, also on a post-Thanksgiving weekend afternoon, following the introduction by our mutual friend Ron Olson, Charlie and I first struck up a deep conversation which ran for several hours. It began an investment partnership that has now endured two decades. Charlie became my mentor, partner, dear friend and above all, life-long role model.

I was so deeply grateful that the Munger family made a special arrangement the next day for me to say a proper and private goodbye to Charlie.

There, lying quietly with eyes closed, Charlie looked the same as ever, peaceful and sincere with a subtle smile on his face. There was a serenity about him. For a moment, I was reminded of the Living Buddhas I once saw in the Buddhist temples of Thailand. In the Buddhist tradition, the bodies of truly enlightened monks, through life-long self-cultivation, can remain incorrupt, without any traces of mummification after death. In that moment, it is what I saw in Charlie, an enlightened sage with an incorruptible body, surrounded by a glimmer of eternal light.

Charlie was not a Buddhist. That vision can never be tested. But it is incontrovertible that his legacy and impact will live on for generations to come.

In our capitalist society, where do virtue, moral responsibility, truth-seeking and public service fit in? Charlie Munger answered these questions through his long exemplary life. He insisted on making money in the most morally sound way, entering transactions only when, if positions were reversed, he would comfortably take the other side. He sought worldly wisdom through life-long learning. He guided life with rationality devoid of mental deficiencies such as envy, resentment and self-pity. He faced and persevered through countless adversities with stoicism and equanimity. As he gained in wealth and stature, he showed little appetite for the trappings of that success, and instead spent his wealth on worthy causes and tirelessly spread his worldly wisdom to those who would listen, often with humor. He remained deeply engaged with family, friends, partners and the broader world with loving assiduousness through his last days.

In his later decades, Charlie Munger’s ideas began to spread across the world, particularly in the most populous countries of China and India. In China, the Mandarin language version of “Poor Charlie’s Almanack,” an anthology by and about Charlie Munger, sold over 1.2 million copies over the last 10 years. There, the educated class increasingly came to view Charlie as the embodiment of the modern-day Confucianism, maintaining a virtuous and enlightened life while embracing the market forces of capitalism. In time, that vision of modern Confucianism will be crucial for Chinese modernization and how China interacts with the rest of the world.

Charlie’s teachings will continue to spread, inspire and impact the world even more profoundly. That will be his eternal legacy.