Telegram Group & Telegram Channel
🧠 DeepSeek ΠΎΠ±Π½Π°Ρ€ΡƒΠΆΠΈΠ»ΠΈ, Ρ‡Ρ‚ΠΎ Ρƒ ΠΈΡ… Π½ΠΎΠ²ΠΎΠΉ ΠΌΠΎΠ΄Π΅Π»ΠΈ Π±Ρ‹Π» ΠΌΠΎΠΌΠ΅Π½Ρ‚ озарСния, ΠΊΠΎΠ³Π΄Π° ΠΎΠ½Π° сама для сСбя Ρ€Π°Π·Ρ€Π°Π±ΠΎΡ‚Π°Π»Π° ΠΏΡ€ΠΎΠ΄Π²ΠΈΠ½ΡƒΡ‚ΡƒΡŽ Ρ‚Π΅Ρ…Π½ΠΈΠΊΡƒ рассуТдСния.

ΠžΠΊΠ°Π·Ρ‹Π²Π°Π΅Ρ‚ΡΡ, Π²Π°ΠΌ просто Π½ΡƒΠΆΠ½ΠΎ ΠΏΡ€Π°Π²ΠΈΠ»ΡŒΠ½ΠΎ ΡΡ‚ΠΈΠΌΡƒΠ»ΠΈΡ€ΠΎΠ²Π°Ρ‚ΡŒ модСль.

Π§ΠΈΡ‚ΠΎΠΉ Π²ΠΎΠ΄Ρ‹ ΠΎΠ±ΡƒΡ‡Π΅Π½ΠΈΠ΅ с ΠΏΠΎΠ΄ΠΊΡ€Π΅ΠΏΠ»Π΅Π½ΠΈΠ΅ΠΌ (RL) ΠΌΠΎΠΆΠ΅Ρ‚ Π½Π°ΡƒΡ‡ΠΈΡ‚ΡŒ модСль Π΄ΡƒΠΌΠ°Ρ‚ΡŒ ΠΈ Ρ€Π΅Ρ„Π»Π΅ΠΊΡΠΈΡ€ΠΎΠ²Π°Ρ‚ΡŒ.

ΠœΡ‹ возвращаСмся Π² эпоху AlphaGo: играя Π² бСсчислСнныС ΠΏΠ°Ρ€Ρ‚ΠΈΠΈ Go ΠΈ максимально увСличивая Ρ„ΡƒΠ½ΠΊΡ†ΠΈΡŽ вознаграТдСния (Π²Ρ‹ΠΈΠ³Ρ€Ρ‹Ρˆ Π² ΠΈΠ³Ρ€Π΅), ΠΈΡΠΏΠΎΠ»ΡŒΠ·ΡƒΡ чистый RL, AlphaGo научился ΠΏΠΎΠ±Π΅ΠΆΠ΄Π°Ρ‚ΡŒ Π»ΡƒΡ‡ΡˆΠΈΡ… ΠΈΠ³Ρ€ΠΎΠΊΠΎΠ² ΠΌΠΈΡ€Π°.

ΠŸΠΎΡ…ΠΎΠΆΠ΅ это Π±ΡƒΠ΄Π΅Ρ‚ эра LLM RL.

πŸ“• Paper

@ai_machinelearning_big_data

#DeepSeek #deepseekr1 #reasoning #ml
Please open Telegram to view this post
VIEW IN TELEGRAM
Please open Telegram to view this post
VIEW IN TELEGRAM



group-telegram.com/ai_machinelearning_big_data/6588
Create:
Last Update:

🧠 DeepSeek ΠΎΠ±Π½Π°Ρ€ΡƒΠΆΠΈΠ»ΠΈ, Ρ‡Ρ‚ΠΎ Ρƒ ΠΈΡ… Π½ΠΎΠ²ΠΎΠΉ ΠΌΠΎΠ΄Π΅Π»ΠΈ Π±Ρ‹Π» ΠΌΠΎΠΌΠ΅Π½Ρ‚ озарСния, ΠΊΠΎΠ³Π΄Π° ΠΎΠ½Π° сама для сСбя Ρ€Π°Π·Ρ€Π°Π±ΠΎΡ‚Π°Π»Π° ΠΏΡ€ΠΎΠ΄Π²ΠΈΠ½ΡƒΡ‚ΡƒΡŽ Ρ‚Π΅Ρ…Π½ΠΈΠΊΡƒ рассуТдСния.

ΠžΠΊΠ°Π·Ρ‹Π²Π°Π΅Ρ‚ΡΡ, Π²Π°ΠΌ просто Π½ΡƒΠΆΠ½ΠΎ ΠΏΡ€Π°Π²ΠΈΠ»ΡŒΠ½ΠΎ ΡΡ‚ΠΈΠΌΡƒΠ»ΠΈΡ€ΠΎΠ²Π°Ρ‚ΡŒ модСль.

Π§ΠΈΡ‚ΠΎΠΉ Π²ΠΎΠ΄Ρ‹ ΠΎΠ±ΡƒΡ‡Π΅Π½ΠΈΠ΅ с ΠΏΠΎΠ΄ΠΊΡ€Π΅ΠΏΠ»Π΅Π½ΠΈΠ΅ΠΌ (RL) ΠΌΠΎΠΆΠ΅Ρ‚ Π½Π°ΡƒΡ‡ΠΈΡ‚ΡŒ модСль Π΄ΡƒΠΌΠ°Ρ‚ΡŒ ΠΈ Ρ€Π΅Ρ„Π»Π΅ΠΊΡΠΈΡ€ΠΎΠ²Π°Ρ‚ΡŒ.

ΠœΡ‹ возвращаСмся Π² эпоху AlphaGo: играя Π² бСсчислСнныС ΠΏΠ°Ρ€Ρ‚ΠΈΠΈ Go ΠΈ максимально увСличивая Ρ„ΡƒΠ½ΠΊΡ†ΠΈΡŽ вознаграТдСния (Π²Ρ‹ΠΈΠ³Ρ€Ρ‹Ρˆ Π² ΠΈΠ³Ρ€Π΅), ΠΈΡΠΏΠΎΠ»ΡŒΠ·ΡƒΡ чистый RL, AlphaGo научился ΠΏΠΎΠ±Π΅ΠΆΠ΄Π°Ρ‚ΡŒ Π»ΡƒΡ‡ΡˆΠΈΡ… ΠΈΠ³Ρ€ΠΎΠΊΠΎΠ² ΠΌΠΈΡ€Π°.

ΠŸΠΎΡ…ΠΎΠΆΠ΅ это Π±ΡƒΠ΄Π΅Ρ‚ эра LLM RL.

πŸ“• Paper

@ai_machinelearning_big_data

#DeepSeek #deepseekr1 #reasoning #ml

BY Machinelearning





Share with your friend now:
group-telegram.com/ai_machinelearning_big_data/6588

View MORE
Open in Telegram


Telegram | DID YOU KNOW?

Date: |

"Someone posing as a Ukrainian citizen just joins the chat and starts spreading misinformation, or gathers data, like the location of shelters," Tsekhanovska said, noting how false messages have urged Ukrainians to turn off their phones at a specific time of night, citing cybersafety. But Kliuchnikov, the Ukranian now in France, said he will use Signal or WhatsApp for sensitive conversations, but questions around privacy on Telegram do not give him pause when it comes to sharing information about the war. In a statement, the regulator said the search and seizure operation was carried out against seven individuals and one corporate entity at multiple locations in Ahmedabad and Bhavnagar in Gujarat, Neemuch in Madhya Pradesh, Delhi, and Mumbai. Friday’s performance was part of a larger shift. For the week, the Dow, S&P 500 and Nasdaq fell 2%, 2.9%, and 3.5%, respectively. Either way, Durov says that he withdrew his resignation but that he was ousted from his company anyway. Subsequently, control of the company was reportedly handed to oligarchs Alisher Usmanov and Igor Sechin, both allegedly close associates of Russian leader Vladimir Putin.
from us


Telegram Machinelearning
FROM American