Search Results for author:
Found 15 papers, 9 papers with code
Neural networks with differentiable structure
We test this method on recurrent neural networks applied to simple sequence prediction problems.
Learning to learn with backpropagation of Hebbian plasticity
As a result, the networks "learn how to learn" in order to solve the problem at hand: the trained networks automatically perform fast learning of unpredictable environmental features during their lifetime, expanding the range of solvable problems.
The impossibility of "fairness": a generalized impossibility result for decisions
Here we show that, when groups differ in prevalence of the predicted event, several intuitive, reasonable measures of fairness (probability of positive prediction given occurrence or non-occurrence; probability of occurrence given prediction or non-prediction; and ratio of predictions over occurrences for each group) are all mutually exclusive: if one of them is equal among groups, the other two must differ.
Differentiable plasticity: training plastic neural networks with backpropagation
How can we build agents that keep learning from experience, quickly and efficiently, after their initial training?
Differentiable Hebbian Consolidation for Continual Learning
Continual learning is the problem of sequentially learning new tasks or knowledge while protecting previously acquired knowledge.
First-Order Preconditioning via Hypergradient Descent
Standard gradient descent methods are susceptible to a range of issues that can impede training, such as high correlations and different scaling in parameter space. These difficulties can be addressed by second-order approaches that apply a pre-conditioning matrix to the gradient to improve convergence.
Learning to Continually Learn
Continual lifelong learning requires an agent or model to learn many sequentially ordered tasks, building on previous knowledge without catastrophically forgetting it.
Estimating Q(s,s') with Deep Deterministic Dynamics Gradients
In this paper, we introduce a novel form of value function, $Q(s, s')$, that expresses the utility of transitioning from a state $s$ to a neighboring state $s'$ and then acting optimally thereafter.
Backpropamine: training self-modifying neural networks with differentiable neuromodulated plasticity
We show that neuromodulated plasticity improves the performance of neural networks on both reinforcement learning and supervised learning tasks.
Enabling Continual Learning with Differentiable Hebbian Plasticity
Continual learning is the problem of sequentially learning new tasks or knowledge while protecting previously acquired knowledge.
Hebbian learning with gradients: Hebbian convolutional neural networks with modern deep learning frameworks
Deep learning networks generally use non-biological learning methods.
Learning to acquire novel cognitive tasks with evolution, plasticity and meta-meta-learning
A hallmark of intelligence is the ability to autonomously learn new flexible, cognitive behaviors - that is, behaviors where the appropriate action depends not just on immediate stimuli (as in simple reflexive stimulus-response associations), but on contextual information that must be adequately acquired, stored and processed for each new instance of the task.
Procedural generation of meta-reinforcement learning tasks
The parametrization allows us to randomly generate an arbitrary number of novel simple meta-learning tasks.
Brain-inspired learning in artificial neural networks: a review
Artificial neural networks (ANNs) have emerged as an essential tool in machine learning, achieving remarkable success across diverse domains, including image and speech generation, game playing, and robotics.
Estimating Q(s,s') with Deterministic Dynamics Gradients
In this paper, we introduce a novel form of a value function, $Q(s, s')$, that expresses the utility of transitioning from a state $s$ to a neighboring state $s'$ and then acting optimally thereafter.
