Study on deep neural networks using branching processes and Mehler's formula.
problem Understanding the mathematical role of activation functions in compositional neural networks.
method Connection between compositional kernels and branching processes via Mehler's formula; new random features algorithm.
result Explicit formulas for eigenvalues of compositional kernels quantify complexity.
Deep networks learn sparse hierarchical features without CoD.
problem Overparameterized deep networks struggle with the curse of dimensionality.
method Norm-constrained neural networks for sparse compositional functions.
result Deep networks can learn sparse hierarchical features efficiently.
Generalization error defines the discriminability and the representation power of a deep model. In this work, we claim that feature space design using deep compositional function plays a significant role in generalization along with explicit and implicit regularizations. Our claims are being established with several im…
Paper simplifies DP composition for adaptive privacy budgets, enabling better privacy and accuracy in deep learning.
problem Tension between efficiency and flexibility in DP composition theorems.
method Rényi Differential Privacy (RDP) for adaptive privacy budgets, proving simpler composition theorem with smaller constants.
result Practical DP composition for adaptive privacy budgets, enabling better privacy and accuracy in deep learning.
Deep learning can learn compositional functions more efficiently by breaking them into stages.
problem Understanding why deep learning performs better than shallow models in learning compositional functions.
method Analyzed learnability of compositional target functions using a three-layer fitting model trained with layer-wise spectral estimators.
result Learning compositional functions can be simplified by breaking them into stages, reducing the complexity of the learning problem.
New ANN method for imputing rounded zeros in compositional data.
problem Imputing missing values in compositional data with rounded zeros.
method Artificial Neural Networks (ANNs) for imputation of compositional data.
result ANNs are competitive or better than conventional methods for imputing rounded zeros.
Standard methods in deep learning for natural language processing fail to capture the compositional structure of human language that allows for systematic generalization outside of the training distribution. However, human learners readily generalize in this way, e.g. by applying known grammatical rules to novel words.…
Recent findings in multi-agent deep learning systems point towards the emergence of compositional languages. These claims are often made without exact analysis or testing of the language. In this work, we analyze the emergent language resulting from two different cooperative multi-agent game with more exact measures fo…
Sharp privacy bounds for sequential analysis of sensitive data.
problem Privacy degradation under sequential analysis of sensitive data.
method Edgeworth expansion in f-differential privacy framework.
result Improved privacy bounds under composition with refined approximation accuracy.
Sublinearly structured DNNs achieve feature learning consistency for compositional functions.
problem Achieving feature-learning and prediction consistency in deep neural networks.
method Sublinearly structured DNNs
result Sublinearly structured DNNs match or surpass wide DNNs in prediction.
CTLNet predicts Shanghai Composite Index with improved accuracy.
problem Predicting Shanghai Composite Index for investors and researchers.
method Proposes CNN-Transformer-LSTM Networks (CTLNet) for multivariate time series forecasting.
result CTLNet outperforms state-of-the-art baselines in Shanghai Composite Index prediction.
Gaussian processes struggle with compositional functions, but deep Gaussian processes can outperform.
problem Gaussian process regression struggles with compositional functions.
method We study information-theoretic lower bounds for posterior contraction rates in Gaussian process regression for a continuous regression model.
result Posterior based on any mean-zero Gaussian process can only recover the truth at a rate strictly slower than the minimax rate for generalized additive functions.
Deep neural networks can approximate complex functions through repeated compositions of a fixed-size ReLU network.
problem Understanding the expressive power of deep neural networks through function compositions.
method Demonstrated the surprising expressive power of repeated compositions of a single fixed-size ReLU network.
result Repeated compositions of a single fixed-size ReLU network can approximate 1-Lipschitz continuous functions on [0,1]d with an error O(r−1/d). Memorizing rare examples helps neural networks generalize better.
problem Improving generalization in deep learning models.
method Theoretical analysis and experiments on neural networks with composition capability.
result Memorizing rare examples can help neural networks make correct predictions on rare test examples.
Recent work has studied the emergence of language among deep reinforcement learning agents that must collaborate to solve a task. Of particular interest are the factors that cause language to be compositional -- i.e., express meaning by combining words which themselves have meaning. Evolutionary linguists have found th…
Describes explaining neurons in deep representations using compositional logical concepts.
problem Interpreting neuron behavior in deep neural networks.
method Identifying compositional logical concepts that closely approximate neuron behavior.
result Compositional explanations provide insights into model performance and allow for adversarial example creation.
Deep-HGP uses Bayesian nonparametric approach for complex data regression.
problem Complex data regression with compositional structures.
method Deep Gaussian processes with a squared-exponential kernel, data-driven lengthscale parameters.
result Posterior distribution optimally recovers unknown true regression curve in terms of quadratic loss.
Deep model tackles claim size modeling with quantile-based regression.
problem Actuarial claim size modeling difficulty with no simple distribution.
method Deep composite regression model with quantile splicing point.
result Deep neural network regression models show superiority over classical approaches.
In this paper, we present C-ADAM, the first adaptive solver for compositional problems involving a non-linear functional nesting of expected values. We proof that C-ADAM converges to a stationary point in O(δ−2.25) with δ being a precision parameter. Moreover, we demonstrate the importance of our resul…
Enhances neural networks with prior knowledge through a composite kernel.
problem Lack of effective methods to incorporate prior knowledge into neural networks.
method Integrates a composite kernel combining a neural network kernel and a GP kernel for modeling known properties.
result Demonstrates superior performance and flexibility of the Implicit Composite Kernel (ICK) on synthetic and real-world data.
In the past few years, off-policy reinforcement learning methods have shown promising results in their application for robot control. Deep Q-learning, however, still suffers from poor data-efficiency and is susceptible to stochasticity in the environment or reward functions which is limiting with regard to real-world a…
The current understanding of deep neural networks can only partially explain how input structure, network parameters and optimization algorithms jointly contribute to achieve the strong generalization power that is typically observed in many real-world applications. In order to improve the comprehension and interpretab…
Model predicts composite structures assembly quality with input uncertainty.
problem Accurate prediction of dimensional deviations and residual stress in composite structures assembly.
method Neural Network Gaussian Process considering input uncertainty.
result NNGPIU model outperforms other methods for nonsmooth, nonlinear responses.
NeSS combines neural and symbolic approaches for better compositional generalization.
problem Lack of compositional generalization in deep learning models.
method NeSS uses a neural network to generate traces, executed by a symbolic stack machine with sequence manipulation.
result Achieves 100% generalization performance across multiple domains.
Edgeworth Accountant calculates privacy loss under differential privacy compositions efficiently.
problem Efficiently computing overall privacy loss under composition of private algorithms.
method Analytical approach using f-differential privacy framework and Edgeworth expansion. result Non-asymptotic (ε,δ)-differential privacy bounds with reduced computational cost. Spatial processes with nonstationary and anisotropic covariance structure are often used when modelling, analysing and predicting complex environmental phenomena. Such processes may often be expressed as ones that have stationary and isotropic covariance structure on a warped spatial domain. However, the warping functi…
We show that deep networks are better than shallow networks at approximating functions that can be expressed as a composition of functions described by a directed acyclic graph, because the deep networks can be designed to have the same compositional structure, while a shallow network cannot exploit this knowledge. Thu…
Compositional structures between parts and objects are inherent in natural scenes. Modeling such compositional hierarchies via unsupervised learning can bring various benefits such as interpretability and transferability, which are important in many downstream tasks. In this paper, we propose the first deep latent vari…
Gaussian processes (GPs) are nonparametric priors over functions. Fitting a GP implies computing a posterior distribution of functions consistent with the observed data. Similarly, deep Gaussian processes (DGPs) should allow us to compute a posterior distribution of compositions of multiple functions giving rise to the…
Deep neural networks trained to inpaint partially occluded images show a deep understanding of image composition and have even been shown to remove objects from images convincingly. In this work, we investigate how this implicit knowledge of image composition can be leveraged for fully self-supervised instance separati…
Enhances DGPs with adaptive RKHS Fourier features for better non-stationary pattern modeling.
problem Capturing complex non-stationary patterns in non-linear dynamical systems.
method Integrates ODE-based RKHS Fourier features into DGPs using convolution operations for adaptive amplitude and phase modulation. Uses a doubly stochastic variational inference framework.
result Improved predictive performance across various regression tasks.
In the domain of algorithmic music composition, machine learning-driven systems eliminate the need for carefully hand-crafting rules for composition. In particular, the capability of recurrent neural networks to learn complex temporal patterns lends itself well to the musical domain. Promising results have been observe…
DeepCoDA provides personalized interpretability for complex health data.
problem Interpreting complex health data, especially compositional data, is challenging.
method DeepCoDA framework for high-dimensional compositional data, personalized interpretability through patient-specific weights.
result DeepCoDA maintains state-of-the-art performance and provides coherent, personalized interpretations.
New algorithm tackles nested bi-level optimization problems for robust feature learning.
problem Nested compositional bi-level optimization problems in machine learning.
method Stochastic approximation algorithms for solving nested compositional bi-level optimization problems without matrix inversions.
result Achieves an ε-stationary solution with an oracle complexity of approximately O_T(1/ε^2).
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 method shows how order of gradient updates impacts stability and convergence in deep learning.
problem Training deep learning models can be unstable and computationally expensive.
method Theoretical analysis and experiments with backward-SGD.
result The order of gradient updates affects stability and convergence, leading to improved performance.
With the increasing complexity of modern power systems, conventional dynamic load modeling with ZIP and induction motors (ZIP + IM) is no longer adequate to address the current load characteristic transitions. In recent years, the WECC composite load model (WECC CLM) has shown to effectively capture the dynamic load re…
This paper studies neural network operators and their convergence properties.
problem Understanding the approximation and convergence of neural network operators.
method Proves density results, convergence estimates, and Voronovskaya-type theorems.
result Establishes quantitative convergence estimates and derives Voronovskaya-type theorems.
Deep Gaussian Processes model functions on DAGs with partially observed data.
problem Reconstructing and inferring from partially observed functions on DAGs with noisy measurements.
method Place priors over functions on DAGs, theoretically study prior-collapse behavior, and offer a structured variational approximation.
result Almost-sure lower bounds on the preservation of input distinctions and interpretability of simulator hierarchies.
New findings show disentangled latent representations are not enough for robust compositional generalization.
problem Deep learning models struggle with compositional generalization, especially in out-of-distribution samples.
method Investigated a 2D Gaussian generation task with fully disentangled inputs, then forced disentangled latent representations into full-dimensional output space.
result Forcing disentangled latent representations into full-dimensional output space enables robust compositional generalization.
We assume that a high-dimensional datum, like an image, is a compositional expression of a set of properties, with a complicated non-linear relationship between the datum and its properties. This paper proposes a factorial mixture prior for capturing latent properties, thereby adding structured compositionality to deep…
New framework explains deep neural networks using variational spline theory.
problem Understanding functions learned by deep neural networks.
method Developed a variational framework and function space.
result Deep ReLU networks are solutions to regularized data fitting problems over the proposed function space.
Mix-n-Match improves uncertainty calibration in deep learning.
problem Post-hoc calibration of machine learning classifiers.
method Ensemble and composition strategies to improve accuracy, efficiency, and expressive power.
result Mix-n-Match strategies achieve better data-efficiency and expressive power while maintaining classification accuracy.
The paper tackles deep learning from dependent data, achieving optimal performance.
problem Deep learning from strongly mixing observations, especially with regularization and optimality.
method Sparse-penalized regularization for deep neural networks, oracle inequality for expected excess risk.
result Deep neural network estimator achieves minimax optimal rate for nonparametric autoregression.
Deep Gaussian Processes (DGPs) combine the expressiveness of Deep Neural Networks (DNNs) with quantified uncertainty of Gaussian Processes (GPs). Expressive power and intractable inference both result from the non-Gaussian distribution over composition functions. We propose interpretable DGP based on approximating DGP …
DFNNs predict non-Euclidean responses from Euclidean predictors.
problem Regression with non-Euclidean responses.
method Deep Fréchet neural networks (DFNNs) approximating conditional Fréchet means.
result DFNNs consistently outperform existing methods in empirical studies.
This research proves guarantees on sequence models' generalization to longer and novel sequences.
problem Generalization to longer sequences and novel token combinations in sequence models.
method Provable guarantees on length and compositional generalization for various sequence models.
result Limited capacity models achieve both length and compositional generalization with diverse training distributions.
The composition of elementary behaviors to solve challenging transfer learning problems is one of the key elements in building intelligent machines. To date, there has been plenty of work on learning task-specific policies or skills but almost no focus on composing necessary, task-agnostic skills to find a solution to …