Charting and Navigating the Space of Solutions for Recurrent Neural Networks
6. prosince 2021
Řečníci
Elia Turner
Sprecher:in · 0 Follower:innen
Kabir Dabholkar
Sprecher:in · 0 Follower:innen
Omri Barak
Sprecher:in · 0 Follower:innen
O prezentaci
In recent years Recurrent Neural Networks (RNNs) were successfully used to model the way neural activity drives task-related behavior in animals, operating under the implicit assumption that the obtained solutions are universal. Observations in both neuroscience and in machine learning challenge this assumption. Animals can approach a given task with a variety of strategies, and training machine learning algorithms introduces the phenomenon of underspecification. These observations imply that every task is associated with a space of solutions. To date, the structure of this space is not understood, limiting the approach of comparing RNNs with neural data.Here, we characterize the space of solutions associated with a given task, and how hyperparameters bias networks within this space. We first study a simple two-neuron network on a task that leads to multiple solutions. We trace the nature of the final solution back to the network’s initial connectivity, and identify discrete dynamical regimes that underly this diversity. We then examine the neuroscience-inspired Ready-Set-Go timing task. We find a rich set of solutions, even under identical hyperparameters. We uncover this variety by testing the trained networks' ability to extrapolate, as a perturbation to a system often reveals hidden structure. Furthermore, we relate extrapolation patterns to specific dynamical objects and to effective algorithms found by the networks. Using a set of features, we define a space of solutions. We find that most networks stay within the vicinity of the solution defined by their initial connectivity. Taken together, our results shed light on the concept of the space of solutions and its uses both in Machine learning and in Neuroscience.In recent years Recurrent Neural Networks (RNNs) were successfully used to model the way neural activity drives task-related behavior in animals, operating under the implicit assumption that the obtained solutions are universal. Observations in both neuroscience and in machine learning challenge this assumption. Animals can approach a given task with a variety of strategies, and training machine learning algorithms introduces the phenomenon of underspecification. These observations imply that ev…
Organizátor
NeurIPS 2021
Konto · 1,9k Follower:innen
O organizátorovi (NeurIPS 2021)
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Baví vás formát? Nechte SlidesLive zachytit svou akci!
Profesionální natáčení a streamování po celém světě.
Sdílení
Doporučená videa
Prezentace na podobné téma, kategorii nebo přednášejícího
Flexible Learning of Sparse Neural Networks via Constrained L0 Regularization
Später ansehen
Juan Ramirez, …
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Uncertainty Quantification for Ptychography using Normalizing Flows
Später ansehen
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Type Inference as Optimization
Später ansehen
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Improving Transferability of Representations via Augmentation-Aware Self-Supervision
Später ansehen
Hankook Lee, …
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Q&A 4
Später ansehen
Felipe Tobar, …
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Automorphic Equivalence-aware Graph Neural Network
Später ansehen
Fengli Xu, …
Ewigspeicher-Fortschrittswert: 0 = 0.0%