ã¯ããã« ABEJAã®ã¢ããã³ãã«ã¬ã³ãã¼ã®ç¬¬ã»ã»ã»ä½çªç®ã ãã»ã»ã»ï¼ï¼ ABEJAã§ã¯ãæ³åæ å½è ããã£ã¦ãã¾ãã å¼è·å£«ã10å¹´ã»ã©ãã£ã¦ãã¾ããããµã¨ãããã¨ã§æ©æ¢°å¦ç¿ã®ä¸çããã£ã¦ã¿ãããªããç¬å¦ã§ãæ°å¦æ¸ãPRMLãéæ¬ãã«ã¹ãã©æ¬ãã°ãããã§ãã¼å çã®æ¬ãèªãã ããã¦ãæ°å¦ããæ©æ¢°å¦ç¿çè«ããPythonã®åå¼·ãããããOJTã§å¦ãã§ã3å¹´ã»ã©RDãã¼ã ãç«ã¡ä¸ãã¦ãæ©æ¢°å¦ç¿ã¢ãã«ã®éçºã»å®è£ ãæè¡èª¿æ»ã«æºãã£ã¦ãã¾ãããä»ã¯ãæ³åé¢ä¿ã®ä»äºãã¡ã¤ã³ã§ãã ä»æ¥ã®ãã¼ãã¯AIã¨å ¬å¹³æ§ã§ãã è²ã ãã¼ãã¯ããã®ã§ãããABEJAã®ä¸ã®äººã®ãå¸æã«ããå ¬å¹³æ§ã«ãã¾ããã å½åã¯ä»ã®ã¢ããã³ãã«ã¬ã³ãã¼ã®è¨äºã¿ããã«ãã³ã¼ãããã£ã±ãæ¸ãã¦ããå ¬å¹³æ§ç¢ºä¿ã®ããã®ãããããªè«æãå®è£ ãã¦å¹æã試ãã¦ã¿ãããå ¬å¹³æ§ã«é¢ããå®ç¾©ã«ã¤ãã¦çè«é¢ã解説ãã¦ã¿ããç³»ã®è¨äºã«ããããã¨æã£ã¦ããã®
OpenAIã¯GPT-3ã®æ¬¡ã®ç 究ãå§ãã¦ãã¾ãï¼ ä¸éãGPT-3ã®ãã¢ã«æ¹§ãä¸ããä¸ï¼OpenAIã¯Scaling Lawã«é¢ãã2æ¬ã®è«æãã²ã£ããã¨å ¬éãã¾ããï¼ Scaling Lawãä¸è¨ã§èª¬æãããªããTransformerã®æ§è½ã¯ãã£ã3ã¤ã®å¤æ°ã®ã¹ãä¹åã«æ¯é ããã¦ãããã¨ãããã®ã§ãï¼ Scaling Lawã¯ããåä½ã ããªãèå³æ·±ã話ã§çµããããããã¾ããï¼å®éã«è±èªåã§ããã¾ã話é¡ã«ããã£ã¦ãã¾ããï¼ãããï¼ãã®æ³åã®æ¬å½ã®åãã«æ°ã¥ãã¦ããç 究è ããã¦ï¼ãªã話é¡ã«ãªããªãã®ãå°æãã¦ãã¾ãï¼ I am curious why people are not talking more about the OpenAI scaling law papers. For me, they seem very significant. What I heard so far:
ãªãªã¼ã¹ãé害æ å ±ãªã©ã®ãµã¼ãã¹ã®ãç¥ãã
ææ°ã®äººæ°ã¨ã³ããªã¼ã®é ä¿¡
j次ã®ããã¯ãã¼ã¯
kåã®ããã¯ãã¼ã¯
lãã¨ã§èªã
eã³ã¡ã³ãä¸è¦§ãéã
oãã¼ã¸ãéã
{{#tags}}- {{label}}
{{/tags}}