Monotone neural networks can approximate and interpolate functions efficiently.
problem Understanding the efficiency and expressiveness of monotone neural networks.
method Solving the monotone interpolation problem using depth-4 networks and comparing size bounds with arbitrary networks.
result Monotone neural networks can approximate and interpolate functions efficiently, but may require exponential size in high dimensions.
The paper addresses monotonicity in machine learning models for fairness and accountability.
problem Ensuring fairness and accountability in transparent machine learning models.
method Study of three types of monotonicity (individual, weak pairwise, strong pairwise) and propose monotonic groves of neural additive models.
result Monotonic groves of neural additive models maintain transparency, accountability, and fairness.
Deep networks show less disorder and more monotonic behavior.
problem Understanding the disorder and monotonicity in deep neural networks.
method Computing frustration and analyzing near-monotonicity in signed graphs of deep networks.
result Deep networks exhibit less disorder and more monotonic behavior than expected.
GD monotonically decreases GFS sharpness in neural networks and scalar models.
problem Oscillatory behavior of loss in GD training.
method Analysis of GFS sharpness and empirical validation.
result GFS sharpness decreases monotonically during GD training.
COMET learns monotonic neural networks by incorporating counterexamples.
problem Enforcing monotonicity constraints in neural networks for real-world tasks.
method Counterexample-guided learning technique for ReLU neural networks.
result COMET achieves state-of-the-art results and improves model quality.
Monotonic Linear Interpolation property in neural networks persists despite non-convexity.
problem Understanding the geometric properties of neural network loss landscapes.
method Tools from differential geometry to analyze the monotonicity of neural network weights.
result Sufficient conditions for the Monotonic Linear Interpolation property under mean squared error.
Bayesian optimization with preference learning using monotonic neural networks.
problem Optimizing complex systems with multiple conflicting objectives.
method Proposes a neural network ensemble for utility surrogate modeling, leveraging monotonicity.
result Demonstrates superior performance compared to existing methods.
Alternative neural network training using monotone variational inequality.
problem Training neural networks efficiently and with guarantees.
method Using monotone variational inequality to solve non-convex problems efficiently.
result Our approach leads to fast convergence and competitive performance compared to traditional methods.
New model outperforms Neural ODEs while being more efficient.
problem Stable convergence and existence guarantees for implicit-depth models.
method Developed Monotone Operator Equilibrium Network (monDEQ) based on monotone operator theory.
result MonDEQ models outperform Neural ODEs and are more computationally efficient.
Monotonic neural networks have recently been proposed as a way to define invertible transformations. These transformations can be combined into powerful autoregressive flows that have been shown to be universal approximators of continuous probability distributions. Architectures that ensure monotonicity typically enfor…
New method calibrates neural network predictions for better reliability.
problem Improper probability estimates from deep networks leading to unreliable predictions.
method Proposes a constrained optimization approach for a monotonic calibration map.
result Achieves state-of-the-art performance across various datasets and models.
Monotonic neural additive models simplify machine learning for credit scoring.
problem Complex machine learning methods make models less transparent and fair.
method Introducing monotonic neural additive models that simplify neural networks while maintaining regulatory compliance.
result Monotonic neural additive models achieve similar accuracy to complex neural networks but are more transparent and fair.
The monotonic linear interpolation in deep networks often leads to plateaus, revealing biases in optimization.
problem Plateaus in the optimization landscape of deep networks during monotonic linear interpolation.
method Investigated monotonic linear interpolation on deep neural networks, focusing on biases in weights and biases.
result Interpolating weights and biases differently can lead to significant differences in loss and accuracy, revealing biases in optimization.
This paper uses neural networks to accurately model competing risks in survival analysis.
problem Ignoring competing risks leads to biased survival estimation in machine learning models.
method The paper introduces constrained monotonic neural networks to model each competing survival distribution.
result The method ensures exact likelihood maximization with reduced computational cost.
SurvSurf predicts first hitting times for intermittent events without monotonic violations.
problem Predicting first hitting times for intermittent events with monotonicity guarantees.
method Partially monotonic neural network for sequential events, incorporating unobserved events.
result SurvSurf outperforms existing models in MSE and IBS metrics.
This work shows MLPs can approximate monotonic functions without bounded activations.
problem Optimizing MLPs with monotonic constraints and bounded activations.
method Generalized theoretical results showing MLPs with non-negative weights and saturating activations are universal approximators.
result MLPs with non-negative weights and saturating activations are universal approximators for monotonic functions.
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.
Fixed points of nonnegative neural networks are analyzed using fixed point theory.
problem Analyzing fixed points in nonnegative neural networks.
method Fixed point theory, nonlinear Perron-Frobenius theory, monotonic and scalable mappings.
result Conditions for the existence of fixed points in nonnegative neural networks are provided.
A new neural network model for ordinal regression.
problem Ordinal regression with non-proportional odds.
method Interpretable neural network for both continuous and discrete responses, training a non-linear neural network as a coefficient function.
result N3POM preserves interpretability while offering flexibility. This work explains how large neural networks generalize well despite overparameterization.
problem Understanding the generalization behavior of large neural networks.
method Theoretical analysis of approximation and generalization errors in regression and classification tasks.
result Deep overparameterized neural networks are statistically consistent across different tasks when regularization is applied.
Neural Q-learning tackles high-dimensional PDEs.
problem Solving high-dimensional PDEs is computationally challenging.
method Adapting Q-learning from reinforcement learning to solve PDEs.
result The neural network approximator converges to the PDE solution as the network width increases.
Gradient descent on neural nets often operates at the Edge of Stability, where loss behavior is complex but loss decreases over time.
problem Understanding the optimization dynamics of neural networks at the Edge of Stability.
method Empirical demonstration of gradient descent behavior in neural network training.
result Gradient descent on neural networks typically occurs at the Edge of Stability, where loss behavior is non-monotonic but loss decreases over time.
We study deep neural networks with polynomial activations, particularly their expressive power. For a fixed architecture and activation degree, a polynomial neural network defines an algebraic map from weights to polynomials. The image of this map is the functional space associated to the network, and it is an irreduci…
Study quantifies information flow in neural networks using relative entropy and RG analogy.
problem Quantifying information flow in deep neural networks.
method Explicit computation of relative entropy in Ising models and feedforward neural networks.
result Monotonic increase of relative entropy to an asymptotic value, confirming connection to c-theorem.
We show that for neural network functions that have width less or equal to the input dimension all connected components of decision regions are unbounded. The result holds for continuous and strictly monotonic activation functions as well as for the ReLU activation function. This complements recent results on approxima…
Enhances early-exit neural networks for anytime classification.
problem Lack of guaranteed prediction quality improvement with longer computation time.
method Post-hoc modification based on Product-of-Experts to enforce conditional monotonicity.
result Achieves conditional monotonicity in prediction quality, enabling anytime classification.
Lockout solves sparse regularization for neural networks.
problem Sparse regularization for neural networks.
method Fast algorithm for finding all solutions to constrained optimization problems for differentiable functions and increasing monotone constraints.
result Sparse solutions are usually superior in accuracy and interpretability.
Survival regression method improves log-likelihood scores.
problem Improper scoring rules in survival regression models.
method SurvivalMonotonic-net (SuMo-net) with monotonic neural networks.
result SuMo-net achieves state-of-the-art log-likelihood scores.
Develops a robust model for skewed and heavy-tailed data in periodontal studies.
problem Skewed and heavy-tailed data in periodontal pocket depth measurements.
method Flexible two-piece scale Student-t error distribution and deep neural network with monotonicity constraints.
result Robust mode-based estimation resistant to outliers with clinical interpretability.
Study reveals how neural network smoothness affects their vulnerability to adversarial attacks.
problem Understanding adversarial vulnerability in deep learning networks.
method Analysis of manifold smoothness and generalization capability of deep neural networks trained with local errors.
result High generalization accuracy requires a fast power-law decay of eigen-spectrum of hidden representations.
Improves GANs training through game theory.
problem Hard training of GANs due to antagonistic networks.
method Rewrote GAN training as a variational inequality and introduced a stochastic relaxed forward-backward algorithm.
result Algorithm converges to an exact solution or a neighborhood of it under monotonicity.
Large GD stepsizes improve margins and speed up training for non-homogeneous networks.
problem Training efficiency and margin improvement in non-homogeneous two-layer networks.
method Investigation of two distinct phases in GD training, showing margin growth and empirical risk decrease.
result Large GD stepsizes lead to faster convergence and improved margins in non-homogeneous networks.
Alignment of neural network representations is influenced by SNR and sample size.
problem Understanding how neural network representations align across different conditions.
method Controlled training of neural networks on perturbed datasets, analyzing alignment and generalization.
result Alignment varies monotonically with SNR but non-monotonically with sample size, with minimal alignment near the interpolation threshold.
Self-referential meta learning avoids explicit optimization by modifying itself.
problem Dependency on human engineering in meta learning algorithms.
method Investigates self-referential meta learning systems that modify themselves without explicit optimization.
result Self-referential neural networks can improve their own modifications without explicit optimization.
Method learns neural network to overestimate reference function with guarantees.
problem Learning a neural network to overestimate a reference function on a given domain.
method Two-step process: constructing Majoring Points and optimizing a neural network.
result The learned neural network overestimates the reference function on the domain.
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.…
Researchers propose a non-monotone quantum natural gradient for quantum systems.
problem Applying natural gradient methods to quantum systems without monotonicity.
method Introducing a non-monotone quantum natural gradient (QNG) and demonstrating its superiority over conventional QNG.
result Non-monotone QNG outperforms conventional QNG in terms of convergence speed.
Study learns a neuron with non-monotonic activation functions.
problem Learning a single neuron with non-monotonic activation functions.
method Gradient descent (GD) with conditions on activation function and input distribution.
result Learnability of non-monotonic activation functions is established without monotonicity assumption.
This paper introduces a novel monotone curve estimation framework based on convex duality.
problem Estimating smooth, continuous, and monotonic curves in data.
method Convex duality and optimal transport theories.
result Established statistical guarantees for monotone curve estimates.
Monotonic differentiable sorting networks improve upon previous methods.
problem Non-monotonicity in differentiable sorting networks.
method Relaxation of conditional swap operations using sigmoid functions to ensure monotonicity.
result Monotonic differentiable sorting networks improve upon previous methods.
Neural optimal transport improves multivariate conformal prediction.
problem Multivariate quantile regression challenges and existing methods ignore joint distribution geometry.
method Combines neural optimal transport with amortized optimization for efficient training and faster inference.
result Constructs tighter and more informative predictive regions for multivariate conformal prediction.
Optimal regularization can prevent the double descent phenomenon in learning models.
problem The double descent phenomenon in learning models, where test performance is non-monotonic in sample size and model size.
method Theoretical and empirical study of optimal ℓ2 regularization for linear regression models and neural networks. result Optimally-tuned ℓ2 regularization achieves monotonic test performance for certain models and mitigates the double descent phenomenon for more general models. New method uses neural maps to efficiently sample lattice QCD distributions.
problem Challenges in sampling Boltzmann distributions of lattice field theories.
method Sparse triangular transport maps exploiting conditional independence structure of lattice graphs.
result Sparse triangular maps achieve efficient sampling with linear time complexity in lattice size.
Synaptic strength can be seen as probability to propagate impulse, and according to synaptic plasticity, function could exist from propagation activity to synaptic strength. If the function satisfies constraints such as continuity and monotonicity, neural network under external stimulus will always go to fixed point, a…
FISAR uses neural networks to optimize safe reinforcement learning with forward-invariant constraints.
problem Safe reinforcement learning with constraints in safety-critical environments.
method Imposing linear constraints on policy parameters' updating dynamics, using a DNN-based optimizer to satisfy these constraints.
result The policy decreases constraint violation and maximizes cumulative reward monotonically.
We simplify neural networks to 3D to study their topological changes.
problem Understanding how neural network layers affect low-dimensional topological invariants.
method Limiting each layer to a width of 3D space, tracking changes in linking numbers.
result ResNets and transformers are equally powerful in changing linking numbers.
We propose learning deep models that are monotonic with respect to a user-specified set of inputs by alternating layers of linear embeddings, ensembles of lattices, and calibrators (piecewise linear functions), with appropriate constraints for monotonicity, and jointly training the resulting network. We implement the l…
There is a growing body of literature showing that deep neural networks are vulnerable to adversarial input modification. Recently this work has been extended from image classification to malware classification over boolean features. In this paper we present several new methods for training restricted networks in this …