Matches in SemOpenAlex for { <https://semopenalex.org/work/W2909034231> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W2909034231 abstract "Liquid State Machine (LSM) is a brain-inspired architecture used for solving problems like speech recognition and time series prediction. LSM comprises of a randomly connected recurrent network of spiking neurons. This network propagates the non-linear neuronal and synaptic dynamics. Maass et al. have argued that the non-linear dynamics of LSMs is essential for its performance as a universal computer. Lyapunov exponent (mu), used to characterize the non-linearity of the network, correlates well with LSM performance. We propose a complementary approach of approximating the LSM dynamics with a linear state space representation. The spike rates from this model are well correlated to the spike rates from LSM. Such equivalence allows the extraction of a metric (tau_M) from the state transition matrix. tau_M displays high correlation with performance. Further, high tau_M system require lesser epochs to achieve a given accuracy. Being computationally cheap (1800x time efficient compared to LSM), the tau_M metric enables exploration of the vast parameter design space. We observe that the performance correlation of the tau_M surpasses the Lyapunov exponent (mu), (2-4x improvement) in the high-performance regime over multiple datasets. In fact, while mu increases monotonically with network activity, the performance reaches a maxima at a specific activity described in literature as the edge of chaos. On the other hand, tau_M remains correlated with LSM performance even as mu increases monotonically. Hence, tau_M captures the useful memory of network activity that enables LSM performance. It also enables rapid design space exploration and fine-tuning of LSM parameters for high performance." @default.
- W2909034231 created "2019-01-25" @default.
- W2909034231 creator A5004348148 @default.
- W2909034231 creator A5047197375 @default.
- W2909034231 creator A5073593093 @default.
- W2909034231 date "2019-01-18" @default.
- W2909034231 modified "2023-09-27" @default.
- W2909034231 title "Predicting Performance using Approximate State Space Model for Liquid State Machines" @default.
- W2909034231 cites W1480485976 @default.
- W2909034231 cites W1867196520 @default.
- W2909034231 cites W1986964674 @default.
- W2909034231 cites W1995682068 @default.
- W2909034231 cites W2025054170 @default.
- W2909034231 cites W2025920171 @default.
- W2909034231 cites W2029647264 @default.
- W2909034231 cites W2076015009 @default.
- W2909034231 cites W2103179919 @default.
- W2909034231 cites W2112698612 @default.
- W2909034231 cites W2138952627 @default.
- W2909034231 cites W2143301704 @default.
- W2909034231 cites W2161355584 @default.
- W2909034231 cites W2170521560 @default.
- W2909034231 cites W2170968634 @default.
- W2909034231 cites W2214525127 @default.
- W2909034231 cites W2411638330 @default.
- W2909034231 cites W2615772065 @default.
- W2909034231 cites W2749445606 @default.
- W2909034231 cites W2907187847 @default.
- W2909034231 cites W2962752743 @default.
- W2909034231 hasPublicationYear "2019" @default.
- W2909034231 type Work @default.
- W2909034231 sameAs 2909034231 @default.
- W2909034231 citedByCount "0" @default.
- W2909034231 crossrefType "posted-content" @default.
- W2909034231 hasAuthorship W2909034231A5004348148 @default.
- W2909034231 hasAuthorship W2909034231A5047197375 @default.
- W2909034231 hasAuthorship W2909034231A5073593093 @default.
- W2909034231 hasConcept C11413529 @default.
- W2909034231 hasConcept C114614502 @default.
- W2909034231 hasConcept C154945302 @default.
- W2909034231 hasConcept C162324750 @default.
- W2909034231 hasConcept C176217482 @default.
- W2909034231 hasConcept C184720557 @default.
- W2909034231 hasConcept C187736073 @default.
- W2909034231 hasConcept C191544260 @default.
- W2909034231 hasConcept C21547014 @default.
- W2909034231 hasConcept C2777052490 @default.
- W2909034231 hasConcept C2780898871 @default.
- W2909034231 hasConcept C33923547 @default.
- W2909034231 hasConcept C41008148 @default.
- W2909034231 hasConcept C48103436 @default.
- W2909034231 hasConcept C52918065 @default.
- W2909034231 hasConceptScore W2909034231C11413529 @default.
- W2909034231 hasConceptScore W2909034231C114614502 @default.
- W2909034231 hasConceptScore W2909034231C154945302 @default.
- W2909034231 hasConceptScore W2909034231C162324750 @default.
- W2909034231 hasConceptScore W2909034231C176217482 @default.
- W2909034231 hasConceptScore W2909034231C184720557 @default.
- W2909034231 hasConceptScore W2909034231C187736073 @default.
- W2909034231 hasConceptScore W2909034231C191544260 @default.
- W2909034231 hasConceptScore W2909034231C21547014 @default.
- W2909034231 hasConceptScore W2909034231C2777052490 @default.
- W2909034231 hasConceptScore W2909034231C2780898871 @default.
- W2909034231 hasConceptScore W2909034231C33923547 @default.
- W2909034231 hasConceptScore W2909034231C41008148 @default.
- W2909034231 hasConceptScore W2909034231C48103436 @default.
- W2909034231 hasConceptScore W2909034231C52918065 @default.
- W2909034231 hasLocation W29090342311 @default.
- W2909034231 hasOpenAccess W2909034231 @default.
- W2909034231 hasPrimaryLocation W29090342311 @default.
- W2909034231 hasRelatedWork W2013664009 @default.
- W2909034231 hasRelatedWork W2038082961 @default.
- W2909034231 hasRelatedWork W2133724036 @default.
- W2909034231 hasRelatedWork W2161355584 @default.
- W2909034231 hasRelatedWork W2170353177 @default.
- W2909034231 hasRelatedWork W2205436741 @default.
- W2909034231 hasRelatedWork W2348882660 @default.
- W2909034231 hasRelatedWork W2466989841 @default.
- W2909034231 hasRelatedWork W2525193548 @default.
- W2909034231 hasRelatedWork W2770340150 @default.
- W2909034231 hasRelatedWork W281268388 @default.
- W2909034231 hasRelatedWork W2949714012 @default.
- W2909034231 hasRelatedWork W2973067119 @default.
- W2909034231 hasRelatedWork W2978500214 @default.
- W2909034231 hasRelatedWork W2989201050 @default.
- W2909034231 hasRelatedWork W2994769837 @default.
- W2909034231 hasRelatedWork W3099937123 @default.
- W2909034231 hasRelatedWork W3108365919 @default.
- W2909034231 hasRelatedWork W3174558417 @default.
- W2909034231 hasRelatedWork W3208806199 @default.
- W2909034231 isParatext "false" @default.
- W2909034231 isRetracted "false" @default.
- W2909034231 magId "2909034231" @default.
- W2909034231 workType "article" @default.