New algorithm trains deep neural networks without global optimization.
problem Training deep neural networks efficiently and without global optimization.
method Uses random complex exponential activation functions and Markov Chain Monte Carlo sampling.
result Consistently attains theoretical approximation rate for residual networks.
Efficient active learning with abstention reduces label complexity exponentially.
problem Achieving high accuracy with minimal labels.
method Developed a computationally efficient active learning algorithm with abstention.
result Achieves polylog(1/ε) label complexity, reducing by an exponential factor.
Active learning refers to the learning protocol where the learner is allowed to choose a subset of instances for labeling. Previous studies have shown that, compared with passive learning, active learning is able to reduce the label complexity exponentially if the data are linearly separable or satisfy the Tsybakov noi…
The problem of active diagnosis arises in several applications such as disease diagnosis, and fault diagnosis in computer networks, where the goal is to rapidly identify the binary states of a set of objects (e.g., faulty or working) by sequentially selecting, and observing, (noisy) responses to binary valued queries. …
New study shows exponential sample growth for ReQU neural networks.
problem Computing neural network approximations from samples is challenging.
method Information-based complexity tools.
result Functions can be approximated by ReQU neural networks at arbitrary rates but require exponentially growing samples.
In the world of big data, large but costly to label datasets dominate many fields. Active learning, a semi-supervised alternative to the standard PAC-learning model, was introduced to explore whether adaptive labeling could learn concepts with exponentially fewer labeled samples. While previous results show that active…
We derive and analyze a new, efficient, pool-based active learning algorithm for halfspaces, called ALuMA. Most previous algorithms show exponential improvement in the label complexity assuming that the distribution over the instance space is close to uniform. This assumption rarely holds in practical applications. Ins…
Rational neural networks approximate functions more efficiently with less depth.
problem Choosing optimal nonlinear activation functions in neural networks.
method Rational activation functions with optimal bounds and efficiency proofs.
result Rational neural networks approximate smooth functions more efficiently than ReLU networks with exponentially smaller depth.
Paper proposes a new activation function to reduce overfitting and large weight update issues.
problem Overfitting and large weight update problems in neural networks.
method Introduces a new activation function called Thresholded Exponential Rectified Linear Units (TERELU).
result TERELU shows better performance in reducing overfitting and large weight update issues compared to other activation functions.
We describe a framework for designing efficient active learning algorithms that are tolerant to random classification noise and are differentially-private. The framework is based on active learning algorithms that are statistical in the sense that they rely on estimates of expectations of functions of filtered random e…
Activation functions influence behavior and performance of DNNs. Nonlinear activation functions, like Rectified Linear Units (ReLU), Exponential Linear Units (ELU) and Scaled Exponential Linear Units (SELU), outperform the linear counterparts. However, selecting an appropriate activation function is a challenging probl…
New algorithm learns halfspaces with membership queries, achieving near optimal label complexity.
problem Learning halfspaces with membership queries.
method Proposed a new algorithm for learning halfspaces with membership queries, proving near optimal label complexity.
result Achieves near optimal label complexity for learning halfspaces.
Analytical solution found for a three-layer network with a specific activation function.
problem Understanding the power of depth in neural networks.
method Found analytical solutions for a three-layer network with a matrix exponential activation function.
result Analytical solutions for equations involving a three-layer network with a matrix exponential activation function.
Derives equations for deep learning biases and weights, showing data complexity reduction.
problem Understanding interpretability in supervised learning.
method Gradient flow equations and dynamical truncation of training data.
result Data complexity reduction at an exponential rate with training.
New insights on eluder dimension for function approximation in machine learning.
problem Complexity measure for online bandits and reinforcement learning with function approximation.
method Study the relationship between eluder dimension and generalized rank for different activation functions.
result Eluder dimension can be exponentially smaller or larger than generalized rank depending on the activation function.
DeepPAMM models complex survival data with deep learning, improving predictive performance.
problem Complex hazard structures in survival analysis with small data sets and censoring.
method Deep learning framework for piecewise exponential models, addressing high-dimensional feature settings.
result DeepPAMM outperforms other machine learning approaches in predictive performance.
This work generalizes bounds on the number of linear regions in CPWL NNs.
problem Determining the number of linear regions in CPWL neural networks is challenging.
method Generalized bounds on the maximal number of linear regions for arbitrary CPWL activation functions.
result Depth significantly increases the number of linear regions, but not exponentially.
PDBAL targets experiments for probabilistic models to maximize insights.
problem Designing experiments to yield valuable insights efficiently.
method Combines user-specified risk function with probabilistic model to adaptively choose designs.
result PDBAL consistently outperforms standard approaches in simulations and real-world drug screen data.
We construct and analyze active learning algorithms for the problem of binary classification with abstention. We consider three abstention settings: \emph{fixed-cost} and two variants of \emph{bounded-rate} abstention, and for each of them propose an active learning algorithm. All the proposed algorithms can work in th…
New algorithms reduce label collection for online prediction with expert advice.
problem Efficiently predicting binary sequences with expert advice using fewer labels.
method Adaptive selective sampling for exponentially weighted forecasters.
result Label complexity scales roughly as the square root of the number of rounds for a scenario with a strictly better expert.
A new algorithm improves efficiency in selecting examples for deep learning.
problem Efficiently choosing multiple examples to mark up for deep learning on large datasets.
method Large BatchBALD algorithm, approximating BatchBALD with reduced computational complexity.
result Comparable quality in selection while significantly reducing computation time, especially for large batches.
We provide new results concerning label efficient, polynomial time, passive and active learning of linear separators. We prove that active learning provides an exponential improvement over PAC (passive) learning of homogeneous linear separators under nearly log-concave distributions. Building on this, we provide a comp…
Method extends option valuation for 2D Lévy models.
problem Valuation of European options under 2-asset infinite-activity Lévy models.
method Developed numerical method extending Wang et al. (2007) for 1D to 2D, using Fourier transform for integral term and semi-Lagrangian theta-method for temporal discretization.
result Favourable second-order convergence for Normal Tempered Stable dynamics.
Improved diffusion models for image synthesis with better training dynamics.
problem Uneven and ineffective training in diffusion models.
method Redesigned network layers to preserve activation, weight, and update magnitudes.
result Significantly better networks at equal computational complexity, improving FID to 1.81.
FMMNN combines sine activations with multi-component, multi-layer structure for high-frequency function approximation.
problem Effective representation and learning of high-frequency features in neural networks.
method Introduces FMMNN with sine-type activations and multi-component, multi-layer structure.
result FMMNN achieves strong accuracy and favorable convergence on oscillatory function-approximation benchmarks.
Study uses ML and statistical models to analyze climate impacts of industrial growth.
problem Understanding and predicting environmental impacts of industrial activities.
method Comparative analysis of ML and statistical models on time series data.
result ML models outperform statistical models in predicting environmental impacts.
We study the complexity of functions computable by deep feedforward neural networks with piecewise linear activations in terms of the symmetries and the number of linear regions that they have. Deep networks are able to sequentially map portions of each layer's input-space to the same output. In this way, deep models c…
New research explores using exponential activation functions in neural networks, achieving convergence with over-parameterization.
problem Achieving neural network convergence with over-parameterization using exponential activation functions.
method Defined a neural function using an exponential activation function, initialized weights with random Gaussian distributions, and used gradient descent to find optimal weights.
result Gradient descent can find a weight matrix such that the neural function's output is within ε of the labels with high probability.
Analytic networks with bounded coefficients can't outperform polynomial approximations.
problem Approximation limits of neural networks with analytic activation functions under coefficient constraints.
method Deterministic analysis using comparison argument and Bernstein-type estimates.
result Networks with analytic activation functions and controlled coefficients cannot outperform classical polynomial approximation rates on non-analytic targets.
Observing prices of European put and call options, we calibrate exponential Lévy models nonparametrically. We discuss the efficient implementation of the spectral estimation procedures for Lévy models of finite jump activity as well as for self-decomposable Lévy models. Based on finite sample variances, confidence inte…
BCD algorithm finds global minima in neural networks.
problem Training deep neural networks to find global minima.
method Block coordinate descent with skip connections and non-negative projection.
result Proves convergence to global minima for strictly monotonic and ReLU activations.
We propose a Laplace approximation that creates a stochastic unit from any smooth monotonic activation function, using only Gaussian noise. This paper investigates the application of this stochastic approximation in training a family of Restricted Boltzmann Machines (RBM) that are closely linked to Bregman divergences.…
Three-hidden-layer neural networks can approximate Hölder continuous functions uniformly with exponential rate.
problem Approximating Hölder continuous functions with neural networks.
method Introduced Floor-Exponential-Step (FLES) networks with three hidden layers.
result Uniform approximation of Hölder continuous functions with an exponential rate.
For any strictly positive martingale S=exp(X) for which X has a characteristic function, we provide an expansion for the implied volatility. This expansion is explicit in the sense that it involves no integrals, but only polynomials in the log strike. We illustrate the versatility of our expansion by computing t…
New bounds on ReLU networks for low-regular functions.
problem Bounding approximation error for ReLU networks on low-regular functions.
method Complexity analysis of Fourier features residual networks to ReLU networks.
result Approximation error bound proportional to target function norm and inversely proportional to network width and depth.
Higher granularity in MoE models boosts expressivity exponentially.
problem Expressivity of Mixture-of-Experts models with varying granularity.
method Comparing models with different numbers of active experts (granularity).
result Exponential separation in network expressivity based on granularity.
A DP method selects best sparse models in high dimensions efficiently.
problem Model selection in high-dimensional sparse linear regression under privacy constraints.
method Differential privacy (DP) with exponential mechanism and Metropolis-Hastings algorithm.
result The method identifies active features quickly under privacy constraints.
Proposes a new information-theoretic framework for analyzing deep neural networks.
problem Difficulty in analyzing deep neural networks using existing theoretical frameworks.
method Introduces an information-theoretic framework with new notions of regret and sample complexity.
result Establishes sample complexity bounds for deep neural networks that are width-independent and linear in depth.
Developing active inference agents for edge devices with limited resources.
problem Creating effective active inference agents on edge devices with limited computational resources.
method Introducing a software toolbox to accelerate the development of active inference agents by non-experts.
result Accelerates the democratization of active inference agents for edge devices.
Market activity scales near a constant of 0.632 in intrinsic time.
problem Understanding the stability of market scaling laws.
method Modeling market directional changes as a memoryless exponential hazard process and identifying the intrinsic time scaling constant.
result The intrinsic time scaling constant is 1−1/e=0.632. Groups on CAT(0) cube complexes grow exponentially uniformly.
problem Uniform exponential growth of groups acting on CAT(0) cube complexes.
method Study groups acting without global fixed points on CAT(0) square complexes.
result Groups with uniform exponential growth or stabilize Euclidean subcomplexes.
It is well-known that the expressivity of a neural network depends on its architecture, with deeper networks expressing more complex functions. In the case of networks that compute piecewise linear functions, such as those with ReLU activation, the number of distinct linear regions is a natural measure of expressivity.…
In this paper we address the problem of pool based active learning, and provide an algorithm, called UPAL, that works by minimizing the unbiased estimator of the risk of a hypothesis in a given hypothesis space. For the space of linear classifiers and the squared loss we show that UPAL is equivalent to an exponentially…
Pure exploration (aka active testing) is the fundamental task of sequentially gathering information to answer a query about a stochastic environment. Good algorithms make few mistakes and take few samples. Lower bounds (for multi-armed bandit models with arms in an exponential family) reveal that the sample complexity …
We investigate active learning with access to two distinct oracles: Label (which is standard) and Search (which is not). The Search oracle models the situation where a human searches a database to seed or counterexample an existing solution. Search is stronger than Label while being natural to implement in many situati…
Recently, self-normalizing neural networks (SNNs) have been proposed with the intention to avoid batch or weight normalization. The key step in SNNs is to properly scale the exponential linear unit (referred to as SELU) to inherently incorporate normalization based on central limit theory. SELU is a monotonically incre…
Optimizes algorithms for non-concave bandit problems.
problem Optimizing algorithms for non-concave bandit problems.
method Unified zeroth-order optimization paradigm.
result Minimax-optimal algorithms in the dimension for low-rank generalized linear bandit problems.
Study near-maturity convergence rates of American put prices in Lévy models.
problem Analyzing convergence rates of optimal exercise prices in Lévy models.
method Examined two settings: jumps of unbounded and bounded variation, deriving near-maturity expansions.
result Near-maturity convergence rate of optimal exercise price is of order √(T-t).