Ron Meir
YOU?
Author Swipe
View article: Identifying dynamic regulation with machine learning using adversarial surrogates
Identifying dynamic regulation with machine learning using adversarial surrogates Open
Biological systems maintain stability of their function in spite of external and internal perturbations. An important challenge in studying biological regulation is to identify the control objectives based on empirical data. Very often the…
View article: Unsupervised Invariant Risk Minimization
Unsupervised Invariant Risk Minimization Open
We propose a novel unsupervised framework for \emph{Invariant Risk Minimization} (IRM), extending the concept of invariance to settings where labels are unavailable. Traditional IRM methods rely on labeled data to learn representations tha…
View article: Unsupervised Translation of Emergent Communication
Unsupervised Translation of Emergent Communication Open
Emergent Communication (EC) provides a unique window into the language systems that emerge autonomously when agents are trained to jointly achieve shared goals. However, it is difficult to interpret EC and evaluate its relationship with na…
View article: Unsupervised Translation of Emergent Communication
Unsupervised Translation of Emergent Communication Open
Emergent Communication (EC) provides a unique window into the language systems that emerge autonomously when agents are trained to jointly achieve shared goals. However, it is difficult to interpret EC and evaluate its relationship with na…
View article: Data-dependent and Oracle Bounds on Forgetting in Continual Learning
Data-dependent and Oracle Bounds on Forgetting in Continual Learning Open
In continual learning, knowledge must be preserved and re-used between tasks, maintaining good transfer to future tasks and minimizing forgetting of previously learned ones. While several practical algorithms have been devised for this set…
View article: Concept-Best-Matching: Evaluating Compositionality in Emergent Communication
Concept-Best-Matching: Evaluating Compositionality in Emergent Communication Open
Artificial agents that learn to communicate in order to accomplish a given task acquire communication protocols that are typically opaque to a human. A large body of work has attempted to evaluate the emergent communication via various eva…
View article: Statistical curriculum learning: An elimination algorithm achieving an oracle risk
Statistical curriculum learning: An elimination algorithm achieving an oracle risk Open
We consider a statistical version of curriculum learning (CL) in a parametric prediction setting. The learner is required to estimate a target parameter vector, and can adaptively collect samples from either the target model, or other sour…
View article: Characterization of the Distortion-Perception Tradeoff for Finite Channels with Arbitrary Metrics
Characterization of the Distortion-Perception Tradeoff for Finite Channels with Arbitrary Metrics Open
Whenever inspected by humans, reconstructed signals should not be distinguished from real ones. Typically, such a high perceptual quality comes at the price of high reconstruction error, and vice versa. We study this distortion-perception …
View article: Distributional Multivariate Policy Evaluation and Exploration with the Bellman GAN
Distributional Multivariate Policy Evaluation and Exploration with the Bellman GAN Open
The recently proposed distributional approach to reinforcement learning (DiRL) is centered on learning the distribution of the reward-to-go, often referred to as the value distribution. In this work, we show that the distributional Bellman…
View article: Identifying Dynamic Regulation with Adversarial Surrogates
Identifying Dynamic Regulation with Adversarial Surrogates Open
Homeostasis, the ability to maintain a stable internal environment in the face of perturbations, is essential for the functioning of living systems. Given observations of a system, or even a detailed model of one, it is both valuable and e…
View article: Meta-Learning Adversarial Bandit Algorithms
Meta-Learning Adversarial Bandit Algorithms Open
We study online meta-learning with bandit feedback, with the goal of improving performance across multiple tasks if they are similar according to some natural similarity measure. As the first to target the adversarial online-within-online …
View article: Emergent Quantized Communication
Emergent Quantized Communication Open
The field of emergent communication aims to understand the characteristics of communication as it emerges from artificial agents solving tasks that require information exchange. Communication with discrete messages is considered a desired …
View article: Perceptual Kalman Filters: Online State Estimation under a Perfect Perceptual-Quality Constraint
Perceptual Kalman Filters: Online State Estimation under a Perfect Perceptual-Quality Constraint Open
Many practical settings call for the reconstruction of temporal signals from corrupted or missing data. Classic examples include decoding, tracking, signal enhancement and denoising. Since the reconstructed signals are ultimately viewed by…
View article: Identifying regulation with adversarial surrogates
Identifying regulation with adversarial surrogates Open
Homeostasis, the ability to maintain a relatively constant internal environment in the face of perturbations, is a hallmark of biological systems. It is believed that this constancy is achieved through multiple internal regulation and cont…
View article: Kalman smoother error bounds in the presence of misspecified measurements
Kalman smoother error bounds in the presence of misspecified measurements Open
The performance of a discrete-time fixed-interval Kalman smoother in the presence of a misspecified measurement equation is considered. An easily calculable numerical bound for the increment in smoothing error energy is provided in an adve…
View article: Emergent Quantized Communication
Emergent Quantized Communication Open
The field of emergent communication aims to understand the characteristics of communication as it emerges from artificial agents solving tasks that require information exchange. Communication with discrete messages is considered a desired …
View article: Identifying Regulation with Adversarial Surrogates
Identifying Regulation with Adversarial Surrogates Open
Homeostasis, the ability to maintain a relatively constant internal environment in the face of perturbations, is a hallmark of biological systems. It is believed that this constancy is achieved through multiple internal regulation and cont…
View article: Integral Probability Metrics PAC-Bayes Bounds
Integral Probability Metrics PAC-Bayes Bounds Open
We present a PAC-Bayes-style generalization bound which enables the replacement of the KL-divergence with a variety of Integral Probability Metrics (IPM). We provide instances of this bound with the IPM being the total variation metric and…
View article: Online Meta-Learning in Adversarial Multi-Armed Bandits
Online Meta-Learning in Adversarial Multi-Armed Bandits Open
We study meta-learning for adversarial multi-armed bandits. We consider the online-within-online setup, in which a player (learner) encounters a sequence of multi-armed bandit episodes. The player's performance is measured as regret agains…
View article: Enhancing Causal Estimation through Unlabeled Offline Data
Enhancing Causal Estimation through Unlabeled Offline Data Open
Consider a situation where a new patient arrives in the Intensive Care Unit (ICU) and is monitored by multiple sensors. We wish to assess relevant unmeasured physiological variables (e.g., cardiac contractility and output and vascular resi…
View article: Metalearning Linear Bandits by Prior Update
Metalearning Linear Bandits by Prior Update Open
Fully Bayesian approaches to sequential decision-making assume that problem parameters are generated from a known prior. In practice, such information is often lacking. This problem is exacerbated in setups with partial information, where …
View article: A Theory of the Distortion-Perception Tradeoff in Wasserstein Space
A Theory of the Distortion-Perception Tradeoff in Wasserstein Space Open
The lower the distortion of an estimator, the more the distribution of its outputs generally deviates from the distribution of the signals it attempts to estimate. This phenomenon, known as the perception-distortion tradeoff, has captured …
View article: Ensemble Bootstrapping for Q-Learning
Ensemble Bootstrapping for Q-Learning Open
Q-learning (QL), a common reinforcement learning algorithm, suffers from over-estimation bias due to the maximization term in the optimal Bellman operator. This bias may lead to sub-optimal behavior. Double-Q-learning tackles this issue by…
View article: Discount Factor as a Regularizer in Reinforcement Learning
Discount Factor as a Regularizer in Reinforcement Learning Open
Specifying a Reinforcement Learning (RL) task involves choosing a suitable planning horizon, which is typically modeled by a discount factor. It is known that applying RL algorithms with a lower discount factor can act as a regularizer, im…
View article: Discount Factor as a Regularizer in Reinforcement Learning
Discount Factor as a Regularizer in Reinforcement Learning Open
Specifying a Reinforcement Learning (RL) task involves choosing a suitable planning horizon, which is typically modeled by a discount factor. It is known that applying RL algorithms with a lower discount factor can act as a regularizer, im…
View article: Option Discovery in the Absence of Rewards with Manifold Analysis
Option Discovery in the Absence of Rewards with Manifold Analysis Open
Options have been shown to be an effective tool in reinforcement learning, facilitating improved exploration and learning. In this paper, we present an approach based on spectral graph theory and derive an algorithm that systematically dis…
View article: Cell-type specific outcome representation in primary motor cortex
Cell-type specific outcome representation in primary motor cortex Open
Adaptive movements are critical to animal survival. To guide future actions, the brain monitors different outcomes, including achievement of movement and appetitive goals. The nature of outcome signals and their neuronal and network realiz…
View article: Optimal Multivariate Tuning with Neuron-Level and Population-Level Energy Constraints
Optimal Multivariate Tuning with Neuron-Level and Population-Level Energy Constraints Open
Optimality principles have been useful in explaining many aspects of biological systems. In the context of neural encoding in sensory areas, optimality is naturally formulated in a Bayesian setting as neural tuning which minimizes mean dec…