Paper develops a dynamic Bayesian approach for active learning that optimizes exploration-exploitation balance.
problem Balancing exploration and exploitation in active learning for unknown functions.
method Develops BHEEM, a Bayesian hierarchical approach with approximate Bayesian computation for sampling trade-off parameters.
result BHEEM achieves at least 21% and 11% improvement over pure exploration and exploitation strategies respectively.
The concept of pure spinor is generalized, giving rise to the notion of pure subspaces, spinorial subspaces associated to isotropic vector subspaces of non-maximal dimension. Several algebraic identities concerning the pure subspaces are proved here, as well as some differential results. Furthermore, the freedom in the…
New algorithm reduces regret in contextual bandits.
problem Minimizing regret in contextual bandits with side information.
method Contextual-Gap algorithm for simple regret minimization.
result Established performance guarantees on simple regret.
Algorithm achieves optimal pricing with minimal exploration for dynamic markets.
problem Optimal pricing in dynamic markets with contextual information.
method Localized exploration-then-commit (LetC) algorithm with pure exploration, refinement, and exploitation stages.
result Achieves minimax optimal, dimension-free regret bound.
New quantum states capture more information, enabling advanced processing tasks.
problem Quantum information processing challenges with limited statistical information.
method Introducing Random-Coefficient Pure States (RCPS) and exploiting their higher-order statistics.
result RCPS provide richer information than density operators, enabling new quantum tasks.
Algorithm identifies optimal stable matching in uncertain two-sided markets.
problem Sequential learning in two-sided markets with unknown preferences.
method Pure exploration approach with elimination-based algorithms exploiting partial preference information.
result Identification of pervasive stable matching for optimal stable matching identification.
New greedy algorithms improve Bayesian optimisation performance.
problem Optimizing continuous functions with exploration vs exploitation trade-offs.
method Introduced two novel ε-greedy acquisition functions and compared them with conventional methods.
result ε-greedy algorithms generally outperform conventional methods, especially in higher dimensions.
This work presents a novel objective function for the unsupervised training of neural network sentence encoders. It exploits signals from paragraph-level discourse coherence to train these models to understand text. Our objective is purely discriminative, allowing us to train models many times faster than was possible …
SSDMs generate quantum states directly, outperforming classical methods.
problem Generating pure-state quantum representations efficiently.
method Score-based generative model on complex projective manifold.
result SSDMs match target pure-state ensembles by orders of magnitude.
A novel approach using graph learning and synthetic long positions for statistical arbitrage in options markets.
problem Exploiting statistical arbitrage opportunities in options markets using machine learning.
method Two-stage graph learning approach: first stage defines a novel prediction target isolating pure arbitrages via synthetic bonds; second stage proposes SLSA positions.
result Statistically significant outperformance of GL baselines and consistent positive returns with an average P&L-contract information ratio of 0.1627.
New scalable GP approximation using Fourier series decomposition.
problem Scalability and accuracy in Gaussian process approximations.
method Harmonic kernel decomposition (HKD) to decompose kernels orthogonally.
result Significantly outperforms standard variational methods in scalability and accuracy.
Algorithm balances online and offline data for linear bandits.
problem Online learning with an offline dataset in linear bandits.
method Proposes a linear bandit algorithm that uses offline data early and increasingly favors exploration as the horizon grows.
result Establishes regret bounds showing competitive performance with both purely online and offline solutions.
We propose a generic, Bayesian, information geometric approach to the exploration--exploitation trade-off in multi-armed bandit problems. Our approach, BelMan, uniformly supports pure exploration, exploration--exploitation, and two-phase bandit problems. The knowledge on bandit arms and their reward distributions is su…
New approach reduces simulator exploitation by improving strategic robustness.
problem Simulator exploitation leading to reality gap between simulation and real-world performance.
method Formulated as a zero-sum minimax game, providing theoretical guarantees and a convergent active data selection algorithm.
result Proves convergence and reduces prediction error in strategically important regions by 1.5-2.2 times.
New approach reduces simulator exploitation by learning robust models.
problem Simulator exploitation leading to reality gap in reinforcement learning.
method Formulated as a zero-sum minimax game between model player and policy player, providing theoretical guarantees and a convergent active data selection algorithm.
result Reduces prediction error in strategically important regions by 1.5-2.2 times and enables near-optimal real-world performance.
New algorithms for risk-averse bandits minimize regret in finite time.
problem Minimizing regret in finite time for bandit problems.
method Proposes two algorithms for selecting the most probable arm with a good risk-return trade-off.
result Upper bound for the minimum number of experiments before commitment to guarantee a bound on regret.
We propose a new integrated method of exploiting model, batch and domain parallelism for the training of deep neural networks (DNNs) on large distributed-memory computers using minibatch stochastic gradient descent (SGD). Our goal is to find an efficient parallelization strategy for a fixed batch size using P process…
Let G be one of the Artin groups of finite type Bn=Cn, and affine type A~n−1 and C~n−1. In this paper, we show that if α and β are elements of G such that αk=βk for some nonzero integer k, then α and β are conjugate in G. For the Artin …
This paper shows how integrating domain knowledge improves ML models for transprecision computing.
problem Improving ML models for transprecision computing with scarce or complex data.
method Injecting domain knowledge into neural networks through additional features, graph-based topology, and regularization schemes.
result ML models with domain knowledge outperform purely data-driven models by around 38%.
We investigate the representation theory of the polynomial core of the quantum Teichmuller space of a punctured surface S. This is a purely algebraic object, closely related to the combinatorics of the simplicial complex of ideal cell decompositions of S. Our main result is that irreducible finite-dimensional represent…
Quantum variational circuits improve reinforcement learning efficiency.
problem Improving reinforcement learning algorithms using quantum computing.
method Investigation of quantum variational circuits for DQN and Double DQN, encoding classical data for quantum circuits.
result Quantum variational circuits can solve reinforcement learning tasks with a smaller parameter space.
The paper presents a new framework for complex Support Vector Regression as well as Support Vector Machines for quaternary classification. The method exploits the notion of widely linear estimation to model the input-out relation for complex-valued data and considers two cases: a) the complex data are split into their …
This paper presents a study of the asymptotic geometry of groups with contracting elements, with emphasis on a subclass of statistically convex-cocompact (SCC) actions. The class of SCC actions includes relatively hyperbolic groups, CAT(0) groups with rank-1 elements and mapping class groups, among others. We exploit a…
Optimizes tensor completion using geodesics on Segre manifolds.
problem Incomplete tensor data in recommender systems and spectroscopy.
method Riemannian conjugate gradient optimization with explicit geodesic expressions.
result Recovery of tensor decomposition from as little as 10% of data.
We point out a new view on slow invariant manifolds (SIM) in dynamical systems which departs from a purely geometric covariant characterization implying coordinate independency. The fundamental idea is to treat the SIM as a well-defined geometric object in phase space and elucidate characterizing geometric properties t…
New algorithm achieves near optimal sample complexity for 1-identification problem.
problem Determining if an arm's mean reward is at least a known threshold with high probability.
method Design of Sequential-Exploration-Exploitation (SEE) algorithm with non-asymptotic analysis.
result Achieves near optimality in sample complexity, matching upper and lower bounds up to a polynomial logarithmic factor.
An emerging way of tackling the dimensionality issues arising in the modeling of a multivariate process is to assume that the inherent data structure can be captured by a graph. Nevertheless, though state-of-the-art graph-based methods have been successful for many learning tasks, they do not consider time-evolving sig…
We consider active learning with logged data, where labeled examples are drawn conditioned on a predetermined logging policy, and the goal is to learn a classifier on the entire population, not just conditioned on the logging policy. Prior work addresses this problem either when only logged data is available, or purely…
Survey on risk-aware multi-armed bandits for better decision-making.
problem Risk measures in multi-armed bandits for better decision-making.
method Review of existing research, definition of risk-aware bandit problems, and algorithms for minimizing regret and identifying best arms.
result Consolidation and summarization of existing research on risk measures in multi-armed bandits.
RL for jump-diffusions applies to financial portfolio selection and option hedging.
problem Optimizing control in systems with jump-diffusion dynamics.
method Entropy-regularized exploratory control with stochastic policies, using existing diffusion algorithms with modifications.
result RL algorithms and parameterizations are invariant to jumps in jump-diffusion systems.
LightSBB-M improves generative diffusion modeling with lower 2-Wasserstein distances.
problem Improving generative diffusion models using Schrödinger Bridge and Bass methods.
method Optimizes SBB transport plan with dual representation and tunable beta parameter.
result Achieves up to 32% improvement in 2-Wasserstein distance on synthetic datasets.
The purpose of this article is to describe connections between the loop space of the 2-sphere, Artin's braid groups, a choice of simplicial group whose homotopy groups are given by modules called Lie(n), as well as work of Milnor, and Habegger-Lin on "homotopy string links". The current article exploits Lie algebras as…
Proposes qPO, a new acquisition strategy for batched Bayesian optimization that maximizes the probability of including the optimum.
problem Efficiently identifying top-performing compounds from a large chemical library.
method qPO (multipoint Probability of Optimality) acquisition strategy that maximizes the probability of including the true optimum.
result Empirical evidence shows that qPO is competitive with and complements other state-of-the-art methods in batched Bayesian optimization.
A new invariant for pure braids is defined and shown not to be trivial.
problem Defining a non-trivial invariant for pure braids.
method Using recoupling theory to define a representation of the pure braid group.
result The defined representation is not trivial.
Characterizes optimal-speed quantum state evolution Hamiltonians.
problem Optimal-speed unitary time evolution of pure and quasi-pure quantum states.
method Construction of the manifold of pure states and isometry with flag manifold, characterization of equigeodesic vectors.
result Hamiltonians generating optimal-speed time evolution are fully characterized by equigeodesic vectors of the flag manifold.
This work optimizes identifying good arms in nonparametric multi-armed bandits.
problem Efficiently identifying arms with high means in nonparametric settings.
method Combining reward-maximizing sampling with a nonparametric sequential test for anytime-valid labeling.
result Achieves minimax optimal stopping times for identifying arms above a threshold.
Efficient algorithm for CMDPs reduces to offline density estimation.
problem Offline learning for CMDPs with horizon H.
method Reduction to offline density estimation, layerwise exploration-exploitation tradeoff.
result First efficient and near-optimal reduction from CMDPs to offline density estimation.
Tr-LinUCB reduces regret in stochastic linear bandits by truncating exploration.
problem Over-exploration in LinUCB leads to suboptimal cumulative regret.
method Proposes Tr-LinUCB, a truncated version of LinUCB.
result Achieves O(dlog(T)) regret with S=Cdlog(T), matching lower bound. The CN matrix of a pure braid projection is characterized and applied.
problem Understanding the structure of CN matrices for braid projections.
method Discussion and characterization of patterns and specific matrices.
result Characterization of CN matrix of a pure 6-braid projection and related matrices.
A spacetime denotes a pure radiation field if its energy momentum tensor represents a situation in which all the energy is transported in one direction with the speed of light. In 1989, Wils and later in 1997 Ludwig and Edgar studied the physical properties of pure radiation metrics, which are conformally related to a …
Corrects earlier work on surface orbifold pure braid groups.
problem Proving a four-term exact sequence for surface orbifold pure braid groups.
method Analyzes surface orbifold pure braid groups for all genus ≥ 1, 2D orientable orbifolds with cone points.
result Proves a four-term exact sequence for surface orbifold pure braid groups.
In this mostly survey paper, we investigate the resonance varieties, the lower central series ranks, and the Chen ranks, as well as the residual and formality properties of several families of braid-like groups: the pure braid groups Pn, the welded pure braid groups wPn, the virtual pure braid groups vPn, as w…
SparseTrain uses dynamic sparsity in training deep neural networks on CPUs.
problem Training deep neural networks efficiently on general-purpose processors.
method Exploits dynamic zeros introduced by ReLU in feature maps and gradients.
result Significantly speeds up training on CPUs, up to 1.51x.
We find finite presentations for the automorphism group of the Artin pure braid group and the automorphism group of the pure braid group associated to the full monomial group.
In this paper it is proved that the pure braided Thompson's group BF admits a bi-order, analog to the bi-order of the pure braid groups.
Summary of pure cactus groups and circle points.
problem Understanding pure cactus groups and configuration spaces.
method Summarizes previous work on the topic.
result Summary of results from previous papers and thesis.
Study automorphisms of pure braid groups on sphere homotopy groups.
problem Understanding automorphisms' effect on sphere homotopy groups.
method Examined Delta-group structure, proved invariance of cycle and boundary groups, computed action for few strands.
result Induced action of all automorphisms of pure braid groups on sphere homotopy groups.
In this paper we introduce the framed pure braid group on n strands of an oriented surface, a topological generalisation of the pure braid group Pn. We give different equivalents definitions for framed pure braid groups and we study exact sequences relating these groups with other generalisations of Pn, usually…