Result of the union of Artificial Neural Networks with memory, Memory-Augmented Neural Networks (MANNs) represent a new frontier in artificial intelligence, thanks to the combination of the increasing learning capacity of neural networks with the possibility to store and retrieve relevant information from memory. Being a trend topic in both the machine learning community and in cognitive neuroscience, the MANNs have been developed independently in opposite directions with various achievements in terms of learning performance and biological plausibility. In the present work, we have explored the capacities of MANNs by comparing the performance of specific models on a sequence of cognitive tasks with an increasing demand of memory dynamics, in such a way to identify advantages and limitations and propose new solutions. Specifically, the study involves two neurally faithful models with internal memory (AuGMEnT and HER) and the current top-performing MANN with external memory, the DNC. In particular, the comparative results showed that AuGMEnT model suffers from memory interference that hampers correct learning of tasks affected by temporal credit assignemnt problem, like 12AX task; on the other side, the DNC network confirmed its excellent performance, but it is penalized by low biological foundations. Afterwards, we proposed three variants of AuGMEnT to overcome its learning limitations: leaky-AuGMEnT, deep AuGMEnT and hierarchical AuGMEnT. Morevoer, the addressing scheme of the DNC model has been simplified to pure content-based approach (R-DNC) and coupled with LRUA access module to restore one-shot learning on classification tasks, like the Omniglot task.
Ottenute dall'unione di reti neurali con un sistema di memoria, le Memory-Augmented Neural Networks (MANNs) rappresentano una nuova frontiera dell'intelligenza artificiale, grazie alla combinazione della crescente capacità di apprendimento delle rete neurali artificiali e della possibilità di salvare e recuperare importanti informazioni dalla memoria. Essendo un argomento estremamente attuale sia nel campo del machine learning che della neuroscienza cognitiva, le MANNs sono state sviluppate in direzioni opposte, raggiungendo diversi traguardi in termini di performance di apprendimento e di plausibilità biologica. Il presente lavoro di tesi esplora le capacità delle MANNs comparando le performance di diversi modelli su una serie di task cognitivi con una crescente richiesta di capacità di memorizzazione, per poi identificare vantaggi e criticità e proporre alternative ai modelli originali. Nello specifico, lo studio si concentra su due modelli a memoria interna con forti basi biologiche (AuGMEnT e HER) e su quella che ad oggi è considerata la miglior rete neurale a memoria esterna, il DNC. In particolare, i risultati comparativi portano alla luce una difficoltà del modello AuGMEnT in merito all'immagazzinamento di sequenze di stimoli in maniera stabile, il che non permette la corretta risoluzione di task affetti dal cosiddetto temporal credit assignment problem, come il task 12AX; al contrario, il modello DNC conferma le sue eccellenti prestazioni di apprendimento in tutti i casi testati, ma è penalizzato da una bassa credibilità dal punto di vista neuro-biologico. Pertanto, sono state proposte tre varianti del modello AuGMEnT per superare le sue debolezze di apprendimento: leaky-AuGMEnT, deep AuGMEnT e hierarchical AuGMEnT. Inoltre, lo schema di scrittura/lettura di DNC è stato ridotto ad un approccio puramente content-based (R-DNC) - più biologicamente verosimile - e successivamente integrato con il modulo LRUA per recuperare la capacità di apprendimento one-shot nel caso di task di classificazione come Omniglot.
Memory-augmented neural networks : enhancing biological plausibility and task-learnability
MARTINOLLI, MARCO
2016/2017
Abstract
Result of the union of Artificial Neural Networks with memory, Memory-Augmented Neural Networks (MANNs) represent a new frontier in artificial intelligence, thanks to the combination of the increasing learning capacity of neural networks with the possibility to store and retrieve relevant information from memory. Being a trend topic in both the machine learning community and in cognitive neuroscience, the MANNs have been developed independently in opposite directions with various achievements in terms of learning performance and biological plausibility. In the present work, we have explored the capacities of MANNs by comparing the performance of specific models on a sequence of cognitive tasks with an increasing demand of memory dynamics, in such a way to identify advantages and limitations and propose new solutions. Specifically, the study involves two neurally faithful models with internal memory (AuGMEnT and HER) and the current top-performing MANN with external memory, the DNC. In particular, the comparative results showed that AuGMEnT model suffers from memory interference that hampers correct learning of tasks affected by temporal credit assignemnt problem, like 12AX task; on the other side, the DNC network confirmed its excellent performance, but it is penalized by low biological foundations. Afterwards, we proposed three variants of AuGMEnT to overcome its learning limitations: leaky-AuGMEnT, deep AuGMEnT and hierarchical AuGMEnT. Morevoer, the addressing scheme of the DNC model has been simplified to pure content-based approach (R-DNC) and coupled with LRUA access module to restore one-shot learning on classification tasks, like the Omniglot task.File | Dimensione | Formato | |
---|---|---|---|
2017_10_Martinolli.pdf
accessibile in internet per tutti
Descrizione: Testo della tesi
Dimensione
8.1 MB
Formato
Adobe PDF
|
8.1 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/135880