Weâve developed a simple meta-learning algorithm called Reptile which works by repeatedly sampling a task, performing stochastic gradient descent on it, and updating the initial parameters towards the final parameters learned on that task. Reptile is the application of the Shortest Descent algorithm to the meta-learning setting, and is mathematically similar to first-order MAML (which is a version
2016å¹´1æ15æ¥ã«è¡ãããNVIDIA Deep Learning Day 2016ã§ã®è¬æ¼ã§ãã ã¨ãããã£ã¢ååä¼ç¤¾ãã¨ã³ã¿ã¼ãã©ã¤ãºãã¸ãã¹äºæ¥é¨ äºï¨ æ¦å£« æ¢ã«ãã£ã¼ãã©ã¼ãã³ã°ã®æ´»ç¨ã¯æ¡ãã£ã¦ãã¾ãããã®ã»ãã·ã§ã³ã§ã¯ä¸çä¸ã®æ§ã ãªäºä¾ããç´¹ä»ãã¾ããã¾ããã¯ã³ã´ãã¿ãã»è¨¼å¸ãABEJAãã²ã¹ãã«ãè¿ããã¦ãæ¥æ¬ã§ã®ãã£ã¼ãã©ã¼ãã³ã°ã®å é²çãªäºä¾ããç´¹ä»ãã¦é ãã¾ãã FACEBOOKã®æå±ããBig Surã¢ã¼ããã¯ãã£ãªã©ããã£ã¼ãã©ã¼ãã³ã°ãæ´»ç¨ããããã®ã·ã¹ãã æ§ç¯ã®ãã¦ãã¦ããMicrosoft AzureãIBM SoftLayer ãªã©ããã£ã¼ãã©ã¼ãã³ã°ã«æé©ãªã¯ã©ã¦ãã®æ´»ç¨ã«ã¤ãã¦ããç´¹ä»ãã¾ãã ããã«ãã£ã¼ãã©ã¼ãã³ã°ã®ã·ã¹ãã æ§ç¯ãè«è² ãã¤ã³ãã°ã¬ã¼ã¿ã¼å社ï¼NTTã³ã ã¦ã§ã¢ãã¯ãã¹ã³ã³ãã¹ãã·ã¹ãã è¨ç»ç 究æããã¯ãã¹ãã¼ã¿ãµã¤ã¨ã³ã¹ã»ãã¼ã±ãã£
ãã£ã¼ãã©ã¼ãã³ã°ãçå¨ãæ¯ãã£ã¦ãã¾ããç§ã®å¨ãã§ã¯æ¨å¹´ããå¤ãèãããããã«ãªããç§ãæ¥çµBPããã®é£è¼ã§æ¨å¹´5æã«Googleã®è²·åããDeep Mind社ã«ã¤ãã¦è§¦ãã¾ãããä»å¹´ã¯ããã«ä»ã¾ã§ãã£ã¼ãã©ã¼ãã³ã°ã«ã¤ãã¦è§¦ãã¦ããªãã£ãã¡ãã£ã¢ã§ã触ããããããã«ãªã£ã¦ãã¾ããããä¾ãã°ãã¤ã±ããã¤ããããå æ¥ãé«ç¥ã§ã話é¡ã«ãªã£ã¦ããã®ã§ããã ç§äºã§ãããä»åº¦æ¹¯å·é¶´ç« ããã®TheWaveã¨ããåå¼·ä¼ã§ã人工ç¥è½ã¨ãã¸ãã¹ã«ã¤ãã¦ä¸æéã»ã©ç»å£ããã¦ããã ããã¨ã«ãªãã¾ãããææã»ããã¼ã¨ãããã¨ã§ãããã£ã¼ã«ãºã¹ããã¸ã®è§£ææ©é¢ã«ã¤ãã¦ã¯ãã®ããã°ã§ã以å触ãã¾ãããããæ©æ¢°ã人éãç½®ãæãããã¿ãããªå¦æ³ã¯100å¹´ä½ã¯è¨ããã¦ãããã¨ã§ãããã解ææ©é¢ããæ©æ¢°å¦ç¿ãã人工ç¥è½ããã·ã³ã®ã¥ã©ãªãã£ã¼ããªã©ãå¼ã³åã¯ã©ãã©ãå¤ãããæµè¡ãå»ããããã¾ãããæè¿ããããã®çºéãç¸ã¾ã£
Emerging technology review ã·ãªã³ã³ãã¬ã¼ããã®å 端æè¡åæã¬ãã¼ããå 端æè¡ãå¦ã³ãæ¥æ¬ä¼æ¥ã®çµå¶æ¦ç¥ã¨è£½åè¨ç»çå®ã«å¯ä¸ãæªæ¥ãçã¿åºããã³ãã£ã¼ä¼æ¥ããã¡ã¦ã³ãã¼ã¨æ¥ããã¤ããã¼ã·ã§ã³èªçã®ææ³ã«è¿«ãã âEmerging Technology Reviewâ is research reports featuring the latest technologies. The articles analyze the new wave of the technologies and explores the Silicon Valley culture and innovations. â featured stories ã·ãªã³ã³ãã¬ã¼ããã®å 端æè¡åæã¬ãã¼ã (ããã°ã¸ã®ãªã³ã¯) â about ã·ãªã³ã³ãã¬ã¼ã§çã¾ããå 端æè¡ã解æ VentureCl
IT Leaders ããã ï¼ ãã¯ããã¸ã¼ä¸è¦§ ï¼ æ¥çåå ï¼ å¸å ´åå ï¼ AIãIoTãããã°ãã¼ã¿ãªã©ç¥ã£ã¦ããã¹ã4ã¤ã®ä¿¯ç°å³ æ¥çåå æ¥çååè¨äºä¸è¦§ã¸ [å¸å ´åå] AIãIoTãããã°ãã¼ã¿ãªã©ç¥ã£ã¦ããã¹ã4ã¤ã®ä¿¯ç°å³ 2015å¹´7æ15æ¥(æ°´)ç°å£ 潤ï¼IT Leadersç·¨éé¨ï¼ ãªã¹ã ç±³å½ã§ã¯ãã³ãã£ã¼ãã£ãã¿ãªã¹ããªã©ã®æè³å®¶ããæ å½åéã®ä¿¯ç°å³ï¼ã©ã³ãã¹ã±ã¼ãï¼ãä½æããå ¬éãã¦ããã±ã¼ã¹ããããåéã®åãã追ã£ããä¼æ¥ååãææ¡ããã«ã¯ç¶æ³ãä¸è¦§ãã¦ææ¡ã§ãã俯ç°å³ããã¨ã¦ãé½åãããããã ã ããããã®ç¨®ã®æ´çã¯æ¥æ¬ã®ITãªã¼ãã¼ã«ã¨ã£ã¦ã大ãã«åèã«ãªãã ä¾ãã°ãMachine Intelligenceï¼http://www.shivonzilis.com/ï¼ãããããã¹ãã¼ããã·ã³åéã®ä¿¯ç°å³ã ããã³ã¢æè¡ã ãã§ãArtificial Intel
人éã¨åãããã«ããããã試è¡é¯èª¤ãç¹°ãè¿ããªãããçµã¿ç«ã¦ä½æ¥ãªã©ã®éå課é¡ãèªãå¦ç¿ãã¦ããã¢ã«ã´ãªãºã ããç±³ã«ãªãã©ã«ãã¢å¤§å¦ãã¼ã¯ã¬ã¼æ ¡ï¼UCãã¼ã¯ã¬ã¼ï¼ããããå¦ç¿ç 究室ã®ãã¼ã ãéçºããã人工ç¥è½ã®ä¸ç¨®ã§ããããã£ã¼ãã©ã¼ãã³ã°ï¼æ·±å±¤å¦ç¿ï¼ãã使ã£ããã®ã ãã®ã¢ã«ã´ãªãºã ã«åºã¥ãã½ãããçµã¿è¾¼ãã ããããã®ãBRETTãã¯ãï¼æ¬¡å 空éã§ã®è©³ããå¨è¾ºç°å¢ãããã°ã©ãã³ã°ãããã¨ãªããä»»åã®æåã¨æå¾ã®ç¶æ ãä¸ãã¦ããã ãã§ãã¬ã´ãããã¯ãçµã¿åãããããæ´æã®ãã³ã¬ã¼ãæ£ã«ç½®ããããããã¡ãã®é£è¡æ©ãçµã¿ç«ã¦ãããæ°´çã®ãã£ãããåãã¦ç· ããããã¨ãã£ããã¾ãã¾ãªä½æ¥ã®ããæ¹ãèªãå¦ç¿ãã¦ãã£ãã¨ããã å·¥å ´ã¨éã£ã¦å®¶åºãäºåæã®ç°å¢ã¯å¸¸ã«å¤åããããã®ãããæ³å®ã·ããªãªã«åããã¦ãããããããã°ã©ãã³ã°ãã¦ããããæ¹ã¯ãæ¡ä»¶ãè¨å¤§ã«ãªãå®ç¨çã§ã¯ãªãã¨ãããããããããããã
ãªãªã¼ã¹ãé害æ å ±ãªã©ã®ãµã¼ãã¹ã®ãç¥ãã
ææ°ã®äººæ°ã¨ã³ããªã¼ã®é ä¿¡
å¦çãå®è¡ä¸ã§ã
j次ã®ããã¯ãã¼ã¯
kåã®ããã¯ãã¼ã¯
lãã¨ã§èªã
eã³ã¡ã³ãä¸è¦§ãéã
oãã¼ã¸ãéã
{{#tags}}- {{label}}
{{/tags}}