The self-organizing map system The self-organizing map system equations: m,(k+)=m(k)+Cx[X-m,(k)] m,(k+1)=m,(k) i≠j 2023/7/9
2023/7/9 The self-organizing map system • The self-organizing map system equations: ( ) ( ) ( ) ( ) ( ) 1 1 j j k k j i i m k m k C X m k m k m k i j + = + − + =
The self-organizing map system The self-organizing map is a unsupervised clustering algorithm. Compared with traditional clustering algorithms,its centroid can be mapped a curve or plain,and it remains topological structure. 2023/7/9
2023/7/9 The self-organizing map system The self-organizing map is a unsupervised clustering algorithm. Compared with traditional clustering algorithms, its centroid can be mapped a curve or plain, and it remains topological structure
2.Signal Hebbian Learning (1)Recency effects and forgetting; (2)Asymptotic correlation encoding: (3)Hebbian correlation decoding. 2023/7/9
2023/7/9 2.Signal Hebbian Learning ⑴ Recency effects and forgetting; ⑵ Asymptotic correlation encoding; ⑶ Hebbian correlation decoding
Signal Hebbian Learning The deterministic first-order signal Hebbian learning law: m,=-m,()+S(飞()S,(y,() (4-132) m,()=m,(0)e'+S,(s)S,(s)edk (4-133) 2023/7/9
2023/7/9 Signal Hebbian Learning The deterministic first-order signal Hebbian learning law: (4-132) (4-133) m m t S x t S y t ij ij i i j j = − + ( ) ( ( )) ( ( )) ( ) ( ) ( ) ( ) 0 0 t t s t m t m e S s S s e ds ij ij i j − − = +
Recency effects and forgetting Hebbian synapses learn an exponentially weighted average of sampled patterns. the forgetting term is -m. The simplest local unsupervised learning law: mi =-my 2023/7/9
2023/7/9 Recency effects and forgetting Hebbian synapses learn an exponentially weighted average of sampled patterns. the forgetting term is . The simplest local unsupervised learning law: −mij m m ij ij = −