arXiv research

Improved greedy 2-coordinate updates for optimization problems with constraints.

Multiple-step lookahead policies have demonstrated high empirical competence in Reinforcement Learning, via the use of Monte Carlo Tree Search or Model Predictive Control. In a recent work \cite{efroni2018beyond}, multiple-step greedy policies and their use in vanilla Policy Iteration algorithms were proposed and analy…

Coordinate descent with random coordinate selection is the current state of the art for many large scale optimization problems. However, greedy selection of the steepest coordinate on smooth problems can yield convergence rates independent of the dimension $n$, and requiring upto $n$ times fewer iterations. In this pap…

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…

A commonly cited inefficiency of neural network training by back-propagation is the update locking problem: each layer must wait for the signal to propagate through the full network before updating. Several alternatives that can alleviate this issue have been proposed. In this context, we consider a simpler, but more e…

Study on rapid policy changes in reinforcement learning.

The article introduces inferential moments for analyzing uncertain multivariable systems.

Learning big data by matrix decomposition always suffers from expensive computation, mixing of complicated structures and noise. In this paper, we study more adaptive models and efficient algorithms that decompose a data matrix as the sum of semantic components with incoherent structures. We firstly introduce "GO decom…

A new principle for optimizer selection improves training speed and performance.

Real Time Dynamic Programming (RTDP) is an online algorithm based on Dynamic Programming (DP) that acts by 1-step greedy planning. Unlike DP, RTDP does not require access to the entire state space, i.e., it explicitly handles the exploration. This fact makes RTDP particularly appealing when the state space is large and…

A novel method for learning Bayesian network structures from decentralized data, balancing privacy and efficiency.

In this paper, we introduce DICOD, a convolutional sparse coding algorithm which builds shift invariant representations for long signals. This algorithm is designed to run in a distributed setting, with local message passing, making it communication efficient. It is based on coordinate descent and uses locally greedy u…

DaRE forests enable efficient data deletion from random forests.

Sampling is a fundamental topic in graph signal processing, having found applications in estimation, clustering, and video compression. In contrast to traditional signal processing, the irregularity of the signal domain makes selecting a sampling set non-trivial and hard to analyze. Indeed, though conditions for graph …

Bayesian approach improves $\varepsilon$-greedy exploration in RL.

Enhances PMD with lookahead to improve RL performance.

New algorithm converges to equilibrium in nonconvex-nonconcave optimization problems without dimension dependence.

In their standard form Gaussian processes (GPs) provide a powerful non-parametric framework for regression and classificaton tasks. Their one limiting property is their $\mathcal{O}(N^{3})$ scaling where $N$ is the number of training data points. In this paper we present a framework for GP training with sequential sele…

This paper investigates trajectory tracking problem for a class of underactuated autonomous underwater vehicles (AUVs) with unknown dynamics and constrained inputs. Different from existing policy gradient methods which employ single actor-critic but cannot realize satisfactory tracking control accuracy and stable learn…

Paper proposes algorithms to accurately identify breakpoints in piecewise regression.

Study on optimizing model updates in performative prediction.

Proposes a Monte-Carlo method for sparse signal reconstruction.

We study a logistic model-based active learning procedure for binary classification problems, in which we adopt a batch subject selection strategy with a modified sequential experimental design method. Moreover, accompanying the proposed subject selection scheme, we simultaneously conduct a greedy variable selection pr…

LoCo learns local representations without end-to-end synchronization, improving performance on complex tasks.

We describe $k$-MLE, a fast and efficient local search algorithm for learning finite statistical mixtures of exponential families such as Gaussian mixture models. Mixture models are traditionally learned using the expectation-maximization (EM) soft clustering technique that monotonically increases the incomplete (expec…

Being able to reach any desired location in the environment can be a valuable asset for an agent. Learning a policy to navigate between all pairs of states individually is often not feasible. An all-goals updating algorithm uses each transition to learn Q-values towards all goals simultaneously and off-policy. However …

Simplifies RL training with fewer techniques, reducing bias and instability.

We consider the fully decentralized machine learning scenario where many users with personal datasets collaborate to learn models through local peer-to-peer exchanges, without a central coordinator. We propose to train personalized models that leverage a collaboration graph describing the relationships between user per…

Fewer data weight updates lead to faster convergence in machine learning models.

A new energy-efficient pruning method for federated learning.

Gradient Boosting Decision Tree (GBDT) are popular machine learning algorithms with implementations such as LightGBM and in popular machine learning toolkits like Scikit-Learn. Many implementations can only produce trees in an offline manner and in a greedy manner. We explore ways to convert existing GBDT implementatio…

Proposes a method to refine PDE-driven high-dimensional rare-event simulation.

Study Whittle index learning algorithms for restless bandits with constant stepsizes.

Fast feature selection for SHM using canonical correlation.

Pruning + fine-tuning reduces model complexity and generalizes well for matrix sensing.

Greedy algorithm achieves sublinear regret for various distributions.

Greedy algorithm optimizes consumption habits with power utility.

SymCircuit learns PC structure via entropy-regularized RL, improving inference efficiency and accuracy.

In this paper, we propose a low-rank approximation method based on discrete least-squares for the approximation of a multivariate function from random, noisy-free observations. Sparsity inducing regularization techniques are used within classical algorithms for low-rank approximation in order to exploit the possible sp…

We present an information-theoretic framework for sequential adaptive compressed sensing, Info-Greedy Sensing, where measurements are chosen to maximize the extracted information conditioned on the previous measurements. We show that the widely used bisection approach is Info-Greedy for a family of $k$-sparse signals b…

The paper improves Gaussian process models for efficient batch optimization.

DSPI connects natural policy gradient to policy iteration, proving global convergence.

Meta-learning algorithm improves AI efficiency by teaching itself.

New distributions allow greedy arm selection in sparse bandit problems.

Simpler ε-greedy with longer action durations improves exploration.

This paper is a follow up to the previous author's paper on convex optimization. In that paper we began the process of adjusting greedy-type algorithms from nonlinear approximation for finding sparse solutions of convex optimization problems. We modified there three the most popular in nonlinear approximation in Banach…

Introduces greedy feature selection for classifier-dependent feature ranking.

Greedy policy maximizes information in unknown linear systems.