Improved exploration methods for reinforcement learning with reduced sample complexity.
problem Challenges in reinforcement learning exploration in unknown environments.
method Proposed game-theoretic and trajectory entropy algorithms with improved sample complexity.
result Established statistical advantage of entropy-regularized MDPs for exploration and reduced sample complexity.
The paper extends entropy maximization to multiscale settings and applies it to neural networks.
problem Achieving optimal risk bounds in neural networks using multiscale entropy.
method Generalizing maximum entropy to multiscale settings and applying it to neural networks.
result The multiscale Gibbs posterior can achieve a smaller excess risk than the single-scale Gibbs posterior in a teacher-student scenario.
We introduce Implicit Policy, a general class of expressive policies that can flexibly represent complex action distributions in reinforcement learning, with efficient algorithms to compute entropy regularized policy gradients. We empirically show that, despite its simplicity in implementation, entropy regularization c…
MAXENT method outperforms ML in sparse data with specific prior correlations.
problem Evaluating MAXENT method's validity limits and comparing it with ML.
method Bayesian decision theory, Dirichlet density, KL distance, regularized maximum likelihood.
result MAXENT can outperform ML in sparse data with specific prior correlations.
In this work, we study the credit assignment problem in reward augmented maximum likelihood (RAML) learning, and establish a theoretical equivalence between the token-level counterpart of RAML and the entropy regularized reinforcement learning. Inspired by the connection, we propose two sequence prediction algorithms, …
Unified framework for maximum entropy RL using Tsallis entropy.
problem Generalizing maximum entropy reinforcement learning with various entropies.
method Tsallis MDPs with Tsallis entropy maximization, controlling entropic index.
result Different entropic indices lead to different optimal policies and exploration tendencies.
A new method for RL with continuous actions improves stability and scalability.
problem Stability and scalability issues in existing RL methods.
method Soft policy gradient with entropy regularization, combined with double sampling for soft Bellman equation.
result Outperforms off-policy prior methods in continuous action RL tasks.
New reinforcement learning algorithms improve policy optimization with entropy regularization.
problem Improving policy optimization in reinforcement learning.
method Soft policy gradient theorem (SPGT) and new policy optimization algorithms.
result New algorithms outperform prior works on various benchmark tasks.
Proposes MEDIC to improve incremental learning by preventing forgetting and intransigence.
problem Challenges of forgetting old knowledge and intransigence on new knowledge in incremental learning.
method Maximum Entropy Regularizer (MER) and DropOut Sampling (DOS) to penalize uncertain knowledge and reduce class imbalance.
result Proposed method 'MEDIC' outperforms state-of-the-art algorithms in accuracy, forgetting, and intransigence.
The paper analyzes the convergence rates of smooth message passing algorithms in entropy-regularized MAP inference.
problem Finding the most likely configuration in graphical models with combinatorial optimization.
method Entropy-regularized linear programming relaxations and smooth message passing algorithms.
result The number of iterations sufficient to recover the true integral MAP solution is determined.
Study improves deep reinforcement learning transfer by applying regularization techniques.
problem Improving generalization in deep reinforcement learning across tasks.
method Proposes regularization techniques including sample elimination, early stopping, and maximum entropy regularized adversarial learning.
result Enhanced robustness and generalization in deep transfer reinforcement learning.
In this paper, we propose a general framework to learn a robust large-margin binary classifier when corrupt measurements, called anomalies, caused by sensor failure might be present in the training set. The goal is to minimize the generalization error of the classifier on non-corrupted measurements while controlling th…
A new algorithm learns diverse policies in reinforcement learning.
problem Learning diverse behaviors in reinforcement learning.
method Proposes Maximum Entropy Diverse Exploration (MEDE) algorithm.
result The set of policies learned by MEDE capture the same modalities as the optimal maximum entropy policy.
Quantum ML predicts data with improved speed and accuracy.
problem Predicting data using maximum likelihood in a quantum setting.
method Quantum states embedding and minimization of quantum relative entropy.
result Unified framework for classical and quantum LLMs with performance guarantees.
This work characterizes optimal multiclass learning with regularization.
problem The empirical risk minimization (ERM) algorithm fails in multiclass learning settings.
method Using one-inclusion graphs (OIGs), the work introduces optimal learning algorithms that relax structural risk minimization and incorporate unsupervised learning.
result An optimal learner is introduced that uses a local regularization function and an unsupervised learning stage to learn the regularizer.
Paper develops MRCs for supervised classification using generalized maximum entropy.
problem Developing robust classifiers for decision problems.
method Generalized maximum entropy principle applied to minimax risk classifiers.
result Learning techniques for determining MRCs with performance guarantees.
A novel RL objective and prioritization framework improve performance and sample-efficiency in multi-goal tasks.
problem Learning diverse goals in multi-goal reinforcement learning.
method Maximum entropy regularization for objective and prioritization framework.
result Promising improvements in performance and sample-efficiency on multi-goal robotic tasks.
In this paper, we propose a general framework to learn a robust large-margin binary classifier when corrupt measurements, called anomalies, caused by sensor failure might be present in the training set. The goal is to minimize the generalization error of the classifier on non-corrupted measurements while controlling th…
In this paper, we present a novel and general framework called {\it Maximum Entropy Discrimination Markov Networks} (MaxEnDNet), which integrates the max-margin structured learning and Bayesian-style estimation and combines and extends their merits. Major innovations of this model include: 1) It generalizes the extant …
Maximum entropy modeling is a flexible and popular framework for formulating statistical models given partial knowledge. In this paper, rather than the traditional method of optimizing over the continuous density directly, we learn a smooth and invertible transformation that maps a simple distribution to the desired ma…
The well known maximum-entropy principle due to Jaynes, which states that given mean parameters, the maximum entropy distribution matching them is in an exponential family, has been very popular in machine learning due to its "Occam's razor" interpretation. Unfortunately, calculating the potentials in the maximum-entro…
The paper shows how policy regularization acts like an adversary to improve robustness.
problem Improving robustness of learned policies in reinforcement learning.
method Using convex duality, the paper characterizes adversarial reward perturbations and provides generalization guarantees.
result Policy regularization acts as an adversary to improve robustness against worst-case reward perturbations.
In many real-world applications, data is not collected as one batch, but sequentially over time, and often it is not possible or desirable to wait until the data is completely gathered before analyzing it. Thus, we propose a framework to sequentially update a maximum margin classifier by taking advantage of the Maximum…
The paper strengthens the classical result of MLE convergence to a Gaussian distribution.
problem The classical result of MLE convergence to a Gaussian distribution.
method Sub-Gaussian concentration and entropic normality of the normalized MLE.
result Entropic central limit theorem for a smoothed version of the estimator.
MGD combines maximum entropy and diffusion methods for efficient sampling.
problem Generating samples from limited information in high dimensions.
method Moment Guided Diffusion (MGD) using stochastic differential equations.
result MGD efficiently samples maximum entropy distributions in finite time.
Incorporating feature selection into a classification or regression method often carries a number of advantages. In this paper we formalize feature selection specifically from a discriminative perspective of improving classification/regression accuracy. The feature selection method is developed as an extension to the r…
Develops statistical framework for resolving reward function ambiguity in inverse reinforcement learning.
problem Non-uniqueness of reward functions in inverse reinforcement learning.
method Entropy regularization combined with least-squares reconstruction of the reward from the soft Bellman residual.
result Least-squares reward function is unique and consistent with the expert policy.
Researchers use Gaussian processes to approximate Lagrange multipliers for Maximum-Entropy distributions.
problem Finding Lagrange multipliers for Maximum-Entropy distributions is computationally challenging.
method Employed Gaussian processes to approximate the Lagrange multipliers as a map of moments. Optimized hyperparameters by maximizing log-likelihood.
result Data-driven Maximum-Entropy closure performs well in approximating non-equilibrium distributions.
Enhances RL by controlling policy stochasticity through trajectory entropy constraints.
problem Non-stationary Q-value estimation and short-sighted entropy tuning in maximum entropy RL.
method Proposes TECRL framework with separate Q-functions for reward and entropy, enforcing a trajectory entropy constraint.
result DSAC-E algorithm achieves higher returns and better stability on OpenAI Gym benchmarks.
ISL algorithm tackles deep exploration efficiently.
problem Deep exploration in reinforcement learning.
method Derives ISL algorithm by augmenting RL objective with a novel regularization term.
result Empirically shows state-of-the-art performance on deep-exploration benchmarks.
The paper presents a method to estimate joint interventional distributions from marginal interventional data.
problem Estimating joint interventional distributions from marginal interventional data.
method The paper extends the Causal Maximum Entropy method to use interventional data and employs Lagrange duality to prove the solution lies in the exponential family.
result The method allows for causal feature selection and inference of joint interventional distributions.
We present a new statistical learning paradigm for Boltzmann machines based on a new inference principle we have proposed: the latent maximum entropy principle (LME). LME is different both from Jaynes maximum entropy principle and from standard maximum likelihood estimation.We demonstrate the LME principle BY deriving …
The need to estimate smooth probability distributions (a.k.a. probability densities) from finite sampled data is ubiquitous in science. Many approaches to this problem have been described, but none is yet regarded as providing a definitive solution. Maximum entropy estimation and Bayesian field theory are two such appr…
We discuss the systemic risk implied by the interbank exposures reconstructed with the maximum entropy method. The maximum entropy method severely underestimates the risk of interbank contagion by assuming a fully connected network, while in reality the structure of the interbank network is sparsely connected. Here, we…
A new IRL model recovers reward and state structure from expert demonstrations.
problem Limitation of classical maximum entropy model in capturing state structure.
method Generalized maximum causal entropy for IRL models.
result Empirically outperforms classical models in recovering reward and state structure.
Bayesian models use hyperparameters to indirectly assign priors, and this work shows how these priors can be derived from maximum entropy principles.
problem Understanding the assumptions and dependencies in Bayesian hierarchical models.
method Demonstrates how canonical distributions and maximum entropy principles can be used to derive marginal priors in hierarchical models.
result Marginal priors in hierarchical models derived from maximum entropy principles have different constraints compared to the original priors.
MIND estimates mutual information from ordinal data without full distributional knowledge.
problem Estimating mutual information from ordinal data with limited data.
method Copula-based maximum-entropy estimation of copula entropies.
result MIND estimator is consistent, data-efficient, and unbounded for any sample size.
The maximum entropy principle can be used to assign utility values when only partial information is available about the decision maker's preferences. In order to obtain such utility values it is necessary to establish an analogy between probability and utility through the notion of a utility density function. According…
Many policy gradient methods are variants of Actor-Critic (AC), where a value function (critic) is learned to facilitate updating the parameterized policy (actor). The update to the actor involves a log-likelihood update weighted by the action-values, with the addition of entropy regularization for soft variants. In th…
The paper develops methods to handle missing data using regularized M-estimation in reproducing kernel Hilbert space.
problem Handling missing data in statistical analysis.
method Kernel ridge regression for imputation and maximum entropy method for propensity score estimation.
result The proposed methods achieve statistical consistency and asymptotic equivalence.
Paper derives the maximum entropy characteristics of a rank order distribution for socio-economic applications.
problem Deriving the maximum entropy characteristics of a rank order distribution for socio-economic applications.
method Maximum entropy framework, deriving the discrete generalized beta distribution under a bivariate utility constraint.
result The discrete generalized beta distribution is a natural maximum entropy distribution under an appropriate bivariate utility constraint.
Paper proposes a policy-search algorithm to learn entropy-maximizing exploration policies in reward-free environments.
problem Reward-free learning in high-dimensional, continuous-control domains.
method Maximum Entropy POLicy optimization (MEPOL) algorithm that maximizes a non-parametric state entropy estimate.
result MEPOL learns a maximum-entropy exploration policy that facilitates learning various reward-based tasks.
A new method learns FCMs from noisy data quickly and robustly.
problem Learning large-scale FCMs with noisy data is time-consuming and lacks robustness.
method Transformed learning problem to convex optimization with least-squares and maximum entropy terms.
result The method is rapid, robust, and performs well on synthetic and real-world datasets.
New method calibrates reference distributions for bounded support.
problem Lack of principled method for bounded-support statistical reference distributions.
method Formulated maximum entropy on projective space of nonnegative measures.
result Prescribed acceptance region uniquely determines deformation parameter.
The paper introduces a new intrinsic reward method for exploration in reinforcement learning.
problem Improving exploration in reinforcement learning agents.
method Intrinsic rewards proportional to the entropy of future state-action features.
result The new objective leads to improved visitation of features within individual trajectories.
Recent contributions have framed linear system identification as a nonparametric regularized inverse problem. Relying on ℓ2-type regularization which accounts for the stability and smoothness of the impulse response to be estimated, these approaches have been shown to be competitive w.r.t classical parametric met…
The problem of determining the joint probability distributions for correlated random variables with pre-specified marginals is considered. When the joint distribution satisfying all the required conditions is not unique, the "most unbiased" choice corresponds to the distribution of maximum entropy. The calculation of t…
Framework for robust RL in continuous control with model misspecification.
problem Model misspecification in reinforcement learning for continuous control.
method Integrates robustness into MPO algorithm through worst-case expected return objective and entropy regularization.
result Robust and soft-robust policies outperform non-robust policies in various domains.