Zobrazeno 1 - 10
of 10
pro vyhledávání: '"Lan, Nur"'
Do architectural differences significantly affect the way models represent and process language? We propose a new approach, based on metric-learning encoding models (MLEMs), as a first step to answer this question. The approach provides a feature-bas
Externí odkaz:
http://arxiv.org/abs/2406.12620
Autor:
Jalouzot, Louis, Sobczyk, Robin, Lhopitallier, Bastien, Salle, Jeanne, Lan, Nur, Chemla, Emmanuel, Lakretz, Yair
We introduce Metric-Learning Encoding Models (MLEMs) as a new approach to understand how neural systems represent the theoretical features of the objects they process. As a proof-of-concept, we apply MLEMs to neural representations extracted from BER
Externí odkaz:
http://arxiv.org/abs/2402.11608
Neural networks offer good approximation to many tasks but consistently fail to reach perfect generalization, even when theoretical work shows that such perfect solutions can be expressed by certain architectures. Using the task of formal language le
Externí odkaz:
http://arxiv.org/abs/2402.10013
Associative memory architectures are designed for memorization but also offer, through their retrieval method, a form of generalization to unseen inputs: stored memories can be seen as prototypes from this point of view. Focusing on Modern Hopfield N
Externí odkaz:
http://arxiv.org/abs/2311.06518
How well do neural networks generalize? Even for grammar induction tasks, where the target generalization is fully known, previous works have left the question open, testing very limited ranges beyond the training set and using different success crit
Externí odkaz:
http://arxiv.org/abs/2308.08253
We train neural networks to optimize a Minimum Description Length score, i.e., to balance between the complexity of the network and its accuracy at a task. We show that networks optimizing this objective function master tasks involving memory challen
Externí odkaz:
http://arxiv.org/abs/2111.00600
We propose a general framework to study language emergence through signaling games with neural agents. Using a continuous latent space, we are able to (i) train using backpropagation, (ii) show that discrete messages nonetheless naturally emerge. We
Externí odkaz:
http://arxiv.org/abs/2005.00110
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
There is a tight, bidirectional connection between the formalism that defines how linguistic knowledge is stored and how this knowledge can be learned. In one direction, the formalism can be mapped onto an evaluation metric that allows the child to c
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::98f0d9736fc1ecb18f206f05a1627c77