This work explores the non-convex optimization in compressive learning and the performance of heuristics.
problem The challenge of learning from compressed representations in compressive learning.
method Numerical simulations of the non-convex optimization landscape and heuristic performance.
result Properties of the non-convex optimization landscape and heuristic performance are explored.
This work shows neural networks can solve non-convex constraints problems.
problem Training neural networks under non-convex constraints.
method Project stochastic gradient descent with no-regret analysis of online learning.
result Overparameterized neural networks achieve near-optimal and near-feasible solutions.
New algorithms optimize non-smooth, non-convex objectives with improved complexity.
problem Optimizing non-smooth, non-convex stochastic objectives.
method Reduction to online learning, applying optimistic online learning techniques.
result Improved complexity for finding (δ,ε)-stationary points. Optimizers find approximate global minima in non-convex problems.
problem Understanding why local methods solve non-convex optimization problems.
method Formalizing the hypothesis that many local minima are approximately global minima.
result Most local minima of practical non-convex objectives are approximately global minima.
SGD's uncertainty quantified in non-convex learning problems.
problem Uncertainty quantification in non-convex learning problems.
method Asymptotic normality of SGD iterates and bias characterization.
result SGD iterates are asymptotically normally distributed around the expected value of the invariant distribution.
New insights into using momentum for non-convex optimization.
problem Improving training of non-convex models like deep neural networks.
method Developed a Lyapunov analysis of SGD with momentum using stochastic primal averaging.
result Precise conditions under which SGD+M outperforms SGD and optimal hyper-parameter schedules.
Improved optimization guarantees for deep learning models with Nesterov acceleration.
problem Optimization in non-convex deep learning landscapes.
method Analysis of Nesterov acceleration in benignly non-convex landscapes.
result Identical guarantees can be obtained in optimization problems with weak geometric assumptions, especially in overparametrized deep learning.
First order methods can take extremely long to find global minima of non-convex functions.
problem Finding global minimizers of non-convex functions.
method Designing a family of non-convex functions and using statistical lower bounds for parameter estimation.
result First order methods can take exponential time to converge to a global minimizer.
In this paper, we consider the problem of learning high-dimensional tensor regression problems with low-rank structure. One of the core challenges associated with learning high-dimensional models is computation since the underlying optimization problems are often non-convex. While convex relaxations could lead to polyn…
New methods improve convergence in non-convex non-smooth learning problems.
problem Sparse learning from high-dimensional data with non-convex, non-smooth regularizers.
method Stochastic proximal gradient methods with arbitrary sampling.
result Independent sampling improves performance over uniform sampling.
Develops a new fairness learning approach for multi-task regression models.
problem Fairness in multi-task regression models with biased datasets.
method Uses rank-based non-parametric independence test (Mann Whitney U statistic) and reformulates as non-convex optimization problem.
result Outperforms state-of-the-art methods on fairness metrics.
New approach for distributed online optimization of non-convex losses with sublinear regret.
problem Regret evaluation and consensus in distributed, multi-agent systems with non-convex losses.
method Composite regret metric and consensus-based online normalized gradient (CONGD) approach for pseudo-convex losses; offline optimization oracle for general non-convex losses.
result First sublinear regret bound for general distributed online non-convex learning.
Paper tackles constrained learning with non-convex losses, overcoming challenges with new approach.
problem Challenges in learning with non-convex losses and statistical constraints.
method Learning in the empirical dual domain, bounding empirical duality gap.
result Established a constrained counterpart to classical learning theory.
Recently non-convex optimization approaches for solving machine learning problems have gained significant attention. In this paper we explore non-convex boosting in classification by means of integer programming and demonstrate real-world practicability of the approach while circumventing shortcomings of convex boostin…
SGD converges to global minimum for certain non-convex functions.
problem Theoretical challenges in optimizing non-convex functions in machine learning.
method Perturbed SGD on a broad class of non-convex functions.
result SGD converges to global minimum for certain non-convex functions.
NOVAS uses adaptive stochastic search for non-convex optimization in deep networks.
problem Non-convex optimization challenges in deep neural networks.
method Adaptive stochastic search for non-convex optimization.
result NOVAS outperforms existing alternatives in a structured prediction task.
Non-convex SGD learns halfspaces with adversarial label noise efficiently.
problem Agnostically learning halfspaces in adversarial label noise settings.
method Non-convex SGD optimization for halfspace learning.
result Non-convex SGD achieves misclassification error close to optimal with adversarial noise.
A vast majority of machine learning algorithms train their models and perform inference by solving optimization problems. In order to capture the learning and prediction problems accurately, structural constraints such as sparsity or low rank are frequently imposed or else the objective itself is designed to be a non-c…
We investigate 1) the rate at which refined properties of the empirical risk---in particular, gradients---converge to their population counterparts in standard non-convex learning tasks, and 2) the consequences of this convergence for optimization. Our analysis follows the tradition of norm-based capacity control. We p…
This paper improves convergence guarantees for SGD algorithms in non-convex smooth functions.
problem Theoretical convergence properties of SGD algorithms for non-convex smooth functions.
method Analysis of SGD algorithms with arbitrary data ordering for non-convex smooth functions.
result Enhanced convergence guarantees for incremental gradient and single shuffle SGD, improving the optimization term of convergence guarantee.
Non-convex sparsity-inducing penalties have recently received considerable attentions in sparse learning. Recent theoretical investigations have demonstrated their superiority over the convex counterparts in several sparse learning settings. However, solving the non-convex optimization problems associated with non-conv…
Unified analysis of multi-attribute graph learning with non-convex penalties.
problem Graph inference from multi-attribute data.
method Penalized log-likelihood objective function with ADMM and local linear approximation.
result Local consistency in support recovery and precision matrix estimation for non-convex penalties.
We consider online learning in an adversarial, non-convex setting under the assumption that the learner has an access to an offline optimization oracle. In the general setting of prediction with expert advice, Hazan et al. (2016) established that in the optimization-oracle model, online learning requires exponentially …
Proposes r2SGLD for efficient constrained exploration in non-convex learning.
problem Stagnation in high-temperature chains of reSGLD in distribution tails.
method r2SGLD: replica exchange with reflection steps in a bounded domain.
result Reflection steps enhance mixing rates with quadratic improvement in domain diameter.
Algorithm-dependent generalization error bounds are central to statistical learning theory. A learning algorithm may use a large hypothesis space, but the limited number of iterations controls its model capacity and generalization error. The impacts of stochastic gradient methods on generalization error for non-convex …
Adaptive momentum method solves non-convex min-max problems.
problem Non-convex min-max optimization problems in training generative adversarial networks.
method Proposes an adaptive momentum algorithm for non-convex min-max optimization.
result Establishes non-asymptotic convergence rates for the proposed algorithm.
New algorithm TUSLA improves learning of non-convex neural networks.
problem Optimizing non-convex loss functions in neural networks with superlinear gradient growth.
method Tamed Unadjusted Stochastic Langevin Algorithm (TUSLA) based on SGLD with taming technology.
result Finite-time guarantees for TUSLA to find approximate minimizers of empirical and population risks.
In this paper we develop proximal methods for statistical learning. Proximal point algorithms are useful in statistics and machine learning for obtaining optimization solutions for composite functions. Our approach exploits closed-form solutions of proximal operators and envelope representations based on the Moreau, Fo…
We introduce a novel algorithm for solving learning problems where both the loss function and the regularizer are non-convex but belong to the class of difference of convex (DC) functions. Our contribution is a new general purpose proximal Newton algorithm that is able to deal with such a situation. The algorithm consi…
This thesis tackles non-convex Bayesian learning via scalable dynamic importance sampling algorithms.
problem Non-convex Bayesian learning problem in deep neural networks.
method Replica exchange Langevin Monte Carlo, control variates method, population-chain replica exchange, scalable dynamic importance sampling.
result Control variates method reduces variance and accelerates convergence in non-convex Bayesian learning.
Paper estimates differences in multi-attribute Gaussian graphical models using non-convex penalties.
problem Estimating differences in multi-attribute Gaussian graphical models with similar structure.
method Penalized D-trace loss function with non-convex (log-sum and SCAD) penalties, proximal gradient descent methods.
result Theoretical analysis and numerical examples support consistency in support recovery and estimation.
We consider online forecasting problems for non-convex machine learning models. Forecasting introduces several challenges such as (i) frequent updates are necessary to deal with concept drift issues since the dynamics of the environment change over time, and (ii) the state of the art models are non-convex models. We ad…
Large over-parametrized models learned via stochastic gradient descent (SGD) methods have become a key element in modern machine learning. Although SGD methods are very effective in practice, most theoretical analyses of SGD suggest slower convergence than what is empirically observed. In our recent work [8] we analyze…
Stochastic gradient descent~(SGD) and its variants have attracted much attention in machine learning due to their efficiency and effectiveness for optimization. To handle large-scale problems, researchers have recently proposed several lock-free strategy based parallel SGD~(LF-PSGD) methods for multi-core systems. Howe…
AGGLIO optimizes non-convex functions with local convexity guarantees.
problem Optimizing non-convex functions with local convexity.
method Stage-wise, graduated optimization technique for locally convex functions.
result Global convergence to the global optimum for non-convex and locally convex objectives.
New optimizer G-AdaGrad improves upon AdaGrad for non-convex machine learning problems.
problem Solving non-convex machine learning problems efficiently.
method Proposes a new optimizer G-AdaGrad and analyzes its convergence using state-space models.
result Empirical results show G-AdaGrad performs better than AdaGrad and Adam.
AEGD optimizes non-convex functions with dynamic energy updates.
problem Optimizing non-convex functions efficiently and robustly.
method Adaptive Gradient Descent (AEGD) with a dynamically updated energy variable.
result AEGD achieves energy-dependent convergence rates for both convex and non-convex objectives.
A meta-learning approach improves the performance of alternating minimization for non-convex optimization problems.
problem Optimizing non-convex problems with multiple variables using alternating minimization.
method Meta-learning based alternating minimization (MLAM) to replace handcrafted updating rules.
result The proposed MLAM method outperforms traditional AM-based methods in various non-convex optimization problems.
Improved generalization bounds for SGD in non-convex learning.
problem Understanding generalization properties of SGD in non-convex settings.
method Introducing Type II perturbed SGD (T2pm-SGD) to analyze generalization error bounds.
result Tighter generalization error bounds for SGD in non-convex learning, especially for sub-Gaussian and bounded loss functions.
Matrix completion is a well-studied problem with many machine learning applications. In practice, the problem is often solved by non-convex optimization algorithms. However, the current theoretical analysis for non-convex algorithms relies heavily on the assumption that every entry is observed with exactly the same pro…
Study accelerates optimization methods in non-convex problems, but doesn't improve the algorithm's performance.
problem Understanding the behavior of momentum-based acceleration methods in non-convex, high-dimensional landscapes.
method Used dynamical mean field theory to describe the average dynamics of heavy-ball momentum and Nesterov acceleration in a non-convex model.
result Accelerated dynamics but did not improve the algorithm's performance with respect to gradient descent.
Binary perceptron's instability linked to replica symmetry breaking.
problem Understanding the relationship between algorithmic instability and replica symmetry breaking in binary perceptron learning.
method Established the connection between algorithmic instability and replica symmetry breaking by comparing the instability condition around the fixed point to the instability for breaking the replica symmetric solution of the free energy function.
result The instability condition around the algorithmic fixed point is identical to the instability for breaking the replica symmetric saddle point solution of the free energy function.
Compact, non-convex curve flows are created.
problem Creating compact, non-convex ancient solutions for curve shortening flow.
method Constructed an ancient solution asymptotic to Yin-Yang curve.
result Compact, non-convex ancient solutions for curve shortening flow are demonstrated.
Analyzes optimal learning rate schedules in high-dimensional non-convex optimization problems.
problem Optimizing high-dimensional non-convex loss landscapes.
method Langevin optimization with learning rate decaying as \(η(t) = t^{-β}\).
result To speed up optimization without getting stuck in saddles, a decay rate \(β < 1\) is optimal, contrary to convex setups where \(β = 1\).
Learning with a {\it convex loss} function has been a dominating paradigm for many years. It remains an interesting question how non-convex loss functions help improve the generalization of learning with broad applicability. In this paper, we study a family of objective functions formed by truncating traditional loss f…
Paper tackles fast convergence for non-convex strongly-concave min-max problems.
problem Non-convex strongly-concave min-max problems in deep learning.
method Proximal stage-based method with PL condition for faster convergence.
result Established fast convergence in primal objective gap and duality gap.
The paper analyzes adaptive algorithms in non-convex optimization landscapes.
problem Analyzing adaptive algorithms in non-convex optimization landscapes.
method Stochastic algorithms with decreasing step-size, considering mini-batches and noise.
result Established almost sure convergence to critical points and minimizers.
New framework for optimizing machine learning risks.
problem Optimizing non-decomposable machine learning objectives.
method Empirical X-risk minimization (EXM) framework with algorithmic techniques.
result Developed algorithms for solving EXM with smooth non-convex objectives.