PulseAugur
LIVE 06:28:16
commentary · [2 sources] · · 한국어(KO) Dwarkesh Patel (@dwarkesh_sp) 카파르티의 코멘트를 인용하며, 지속학습을 '가중치와 컨텍스트의 경계를 없애는 것'으로 보는 관점에 반박합니다. 인간은 그런 방식이 아니라, 작업 기억은 지워지고 장기 기억은 더 안정적으로 유지된다는 점을 상기시키며 연속 학습 설계에
1
commentary

Filmmaker Aronofsky sees AI as creator empowerment, Patel critiques continuous learning

By PulseAugur Editorial · Summary by gemini-2.5-flash-lite from 2 sources

Filmmaker Darren Aronofsky believes AI will empower creators rather than replace them, allowing for greater artistic freedom. Dwarkesh Patel, referencing Kaparthy's views, challenges the idea of continuous learning as merely blurring weights and context. Patel emphasizes that human memory functions differently, with working memory fading while long-term memory solidifies, suggesting a more nuanced approach to designing continuous learning systems. AI

Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →

IMPACT AI is seen as a tool to enhance human creativity, while continuous learning models are being re-evaluated for their alignment with human cognitive processes.

RANK_REASON The cluster contains opinions and commentary from a filmmaker and a podcaster on AI's role in creativity and the nature of continuous learning, rather than a direct release or research finding.

Read on Mastodon — fosstodon.org →

COVERAGE [2]

  1. Mastodon — fosstodon.org TIER_1 한국어(KO) · [email protected] ·

    Mark Gadala-Maria (@markgadala) sees AI not as a replacement for creators, but as something that will free storytellers, calling him a rare director who understands the essence of filmmaking. The mention of Orson Welles is also intentional,

    Mark Gadala-Maria (@markgadala) Darren Aronofsky는 AI가 창작자를 대체하는 것이 아니라 오히려 이야기꾼을 더 자유롭게 해줄 것이라고 보며, 영화 창작의 본질을 이해하는 드문 감독이라고 평가한다. Orson Welles를 언급한 맥락도 의도적이며, AI가 예술가의 상상력을 확장할 수 있다는 관점이 핵심이다. https:// x.com/markgadala/status/205583 1490978554283 # ai # storytelling # creativeai…

  2. Mastodon — fosstodon.org TIER_1 한국어(KO) · [email protected] ·

    Citing Dwarkesh Patel (@dwarkesh_sp)'s comment, refutes the view that continuous learning is 'erasing the boundaries of weights and context'. Reminds that humans do not learn that way, but rather working memory is erased and long-term memory is more stably maintained, in the design of continuous learning.

    Dwarkesh Patel (@dwarkesh_sp) 카파르티의 코멘트를 인용하며, 지속학습을 '가중치와 컨텍스트의 경계를 없애는 것'으로 보는 관점에 반박합니다. 인간은 그런 방식이 아니라, 작업 기억은 지워지고 장기 기억은 더 안정적으로 유지된다는 점을 상기시키며 연속 학습 설계에 시사점을 줍니다. https:// x.com/dwarkesh_sp/status/20557 71242620469586 # continuallearning # llm # memory # finetuning # ai

RELATED ENTITIES

RELATED TOPICS