TCNs can approximate complex input-output maps with limited memory.
problem Approximating complex input-output maps with limited memory.
method Proved TCNs can approximate a wide class of input-output maps with arbitrary error tolerance.
result Deep ReLU TCNs can approximate input-output maps with finite memory to arbitrary error.
Safe neural networks for input-output specifications.
problem Ensuring machine learning models adhere to input-output constraints.
method Designing constrained predictors and combining them safely.
result Demonstrated on synthetic and real-world datasets.
URNNs are as expressive as general RNNs with ReLU activations.
problem Expressiveness of URNNs compared to general RNNs.
method Input-output equivalence between URNNs and contractive RNNs with ReLU activations.
result URNNs are as expressive as general RNNs with ReLU activations.
Economic systems, traditionally analyzed as almost independent national systems, are increasingly connected on a global scale. Only recently becoming available, the World Input-Output Database (WIOD) is one of the first efforts to construct the multi-regional input-output (MRIO) tables at the global level. By viewing t…
This paper investigates how economic shocks propagate and amplify through the input-output network connecting industrial sectors in developed economies. We study alternative models of diffusion on networks and we calibrate them using input-output data on real-world inter-sectoral dependencies for several European count…
New approach ties loss curvature to model performance in deep learning.
problem Understanding the relationship between loss curvature and model performance in deep learning.
method Empirical analysis of loss Hessians and theoretical results on input-output Jacobians.
result Novel generalization bound in terms of empirical Jacobian.
Develops neural network approximations for infinite-dimensional input-output maps.
problem Approximating input-output maps between infinite-dimensional spaces.
method Combines neural networks and model reduction techniques.
result Proves convergence of the proposed approximation methodology.
Optimal intervention in economic networks modeled as influence maximization, with hard computational problems.
problem Optimal intervention in economic networks modeled as influence maximization.
method Transformed into influence maximization-like form, with theoretical and practical implications.
result Optimal intervention is NP-hard and cannot be approximated to a constant factor in polynomial time.
Graph-to-Tree Neural Networks improve structured input-output translation in tasks like semantic parsing and math word problems.
problem Improving performance on tasks like semantic parsing and math word problem solving.
method Graph-to-Tree Neural Networks, consisting of a graph encoder and a hierarchical tree decoder.
result Graph2Tree model outperforms or matches state-of-the-art models on neural semantic parsing and math word problem tasks.
Paper presents a method to accurately quantify neural network uncertainty without sampling.
problem Uncertainty quantification in neural networks for reliability and robustness.
method Sample-free moment propagation technique for mean vectors and covariance matrices.
result Analytic solution for covariance of nonlinear activation functions.
Tensor completion method identifies nonlinear systems from input-output data.
problem Identifying nonlinear functions from input-output data pairs.
method Formulated as tensor completion problem with smoothness regularization and solved using block coordinate descent.
result Provable correct nonlinear system identification under certain conditions.
Maxout networks study gradients and propose initialization strategies.
problem Complexity in input-output Jacobian distribution complicates stable parameter initialization.
method Obtained bounds on moments of gradients and formulated initialization strategies.
result Parameter initialization strategies improve training of deep maxout networks.
Paper proposes a method to interpret neural networks by decomposing them into simpler tasks.
problem Understanding the complex nonlinear relationships in trained neural networks.
method Non-negative matrix factorization applied to a trained layered neural network.
result Reveals the roles of hidden units in terms of their contribution to each principal task.
We consider the problem of training input-output recurrent neural networks (RNN) for sequence labeling tasks. We propose a novel spectral approach for learning the network parameters. It is based on decomposition of the cross-moment tensor between the output and a non-linear transformation of the input, based on score …
We compress large neural networks for quick adaptation to specific contexts.
problem How to quickly adapt a pretrained large neural network to specific contexts.
method Propose a Bayesian hypernetwork framework to compress the network and encourage sparsity.
result Generated compressed networks are significantly smaller than baseline methods.
New framework improves robustness of implicit neural networks.
problem Ill-posedness and convergence instability in implicit neural networks.
method NEMON framework based on contraction theory for ℓ∞ norm, including well-posedness condition, average iteration, and input-output Lipschitz constant regularization. result Improved accuracy and robustness of implicit models with smaller input-output Lipschitz bounds.
Linear Memory Network separates memory and function in RNNs.
problem Complex transduction problems requiring memory and input-output exploitation.
method Conceptual separation between memory and function, using feedforward and autoencoder components.
result Efficient training and competitive performance on polyphonic music datasets.
Method combines clustering and matrix completion for missing data in I/O tables.
problem Reconstructing missing entries in World Input-Output (I/O) matrices due to data collection issues.
method Hierarchical clustering and Matrix Completion with LASSO-like nuclear norm penalty.
result The method effectively predicts missing values from previous and similar countries' data.
Wide and shallow networks approximate convex functions well.
problem Understanding why wide and shallow neural networks perform well.
method Analyzing the epigraph of the input-output map of shallow and wide neural networks.
result The epigraph of the input-output map approximates a convex function.
JacNet learns Jacobians to enforce structure on derivatives for invertibility and Lipschitz functions.
problem Enforcing structure on derivatives of neural network mappings.
method Proposes using a neural network to directly learn the Jacobian of the input-output function, allowing control over derivative structure.
result Demonstrates learning invertible approximations to simple and 1-Lipschitz functions.
Procedure removes training data dependency from deep networks, improving generalization.
problem Removing dependency on training data in deep networks for better generalization.
method Deterministic and stochastic parts to ensure forgetting, leveraging activation and weight dynamics.
result New bound on information extraction from black-box networks, ensuring forgetting in activations.
We consider the learning of algorithmic tasks by mere observation of input-output pairs. Rather than studying this as a black-box discrete regression problem with no assumption whatsoever on the input-output mapping, we concentrate on tasks that are amenable to the principle of divide and conquer, and study what are it…
Sharp limits found for storing and retrieving input-output associations in linear associative memories.
problem Understanding the fundamental limits of storing and retrieving input-output associations in neural networks.
method Study of a minimal linear associative memory model, introducing a decoupled model and using statistical physics to characterize storage capacity.
result Linear associative memory can store up to 1/2 log(p) associations, providing a sharp statistical-physics characterization.
Generates long programs from inputs, optimizing multiple tasks.
problem Creating long programs from input-output pairs.
method Trains a neural network to map state and outputs to next program statement, optimizing multiple tasks concurrently.
result Creates programs twice as long as existing solutions, improving success rate and runtime.
Deep neural networks undergo hierarchical free-energy landscape transitions with increasing data size.
problem Understanding the design space and dynamics of deep neural networks.
method Statistical mechanical approach based on replica method.
result Hierarchical free-energy landscape transitions with ultrametricity, leading to simpler configurations in deeper layers.
New method ranks sectors and countries using local and aggregate I-O data.
problem Ranking sectors and countries in global value chains using incomplete I-O tables.
method Rank-1 approximation to I-O tables using local and aggregate information. result Consistently good performance in reconstructing rankings of upstreamness and downstreamness.
NAIS-Net stabilizes deep networks using non-autonomous dynamical systems.
problem Stabilizing deep neural networks to prevent vanishing/exploding gradients.
method NAIS-Net uses non-autonomous dynamical systems with skip connections to enforce stability.
result NAIS-Net proves to be globally asymptotically stable and reduces generalization gap.
This study reveals a walnut-shaped structure in the Japanese production network.
problem The validity and accuracy of conventional input-output analysis in Japanese production networks.
method Infomap method for community detection.
result Most irreducible communities are at the second level of the walnut structure.
Investigates the impact of finite VC dimension on neural network approximation and learning.
problem The influence of VC dimension on neural network approximation and learning from samples.
method Analysis of high-dimensional geometry and statistical learning theory, focusing on VC dimension.
result Finite VC dimension is beneficial for uniform convergence of empirical errors but not for approximation of functions from a probability distribution.
The initial theoretical connections between Leontief input-output models and Markov chains were established back in 1950s. However, considering the wide variety of mathematical properties of Markov chains, there has not been a full investigation of evolving world economic networks with Markov chain formalism. Using the…
Extensions to given-data Sobol' index estimators for large models.
problem Efficiently compute Sobol' indices for models with many inputs.
method General definition, streaming algorithm, heuristic filtering.
result Comparable accuracy and lower memory usage for large models.
Empirical study finds large neural networks are more robust to input perturbations and generalize better.
problem Tension between model complexity and generalization in neural networks.
method Empirical exploration of sensitivity to input perturbations in thousands of models across various architectures and datasets.
result Robustness to input perturbations correlates with generalization in neural networks.
Recursive sketches summarize deep networks, aiding quick analysis and learning.
problem Understanding and analyzing complex deep learning models.
method Developed a recursive sketch mechanism to summarize inputs and outputs of modular deep networks.
result Sketches can identify key components and summarize essential information, even if partially erased.
New approach generates better synthetic data for neural program synthesis.
problem Current approaches to neural program synthesis generalize poorly to real data.
method Adversarial approach to control synthetic data distributions.
result Proposed method outperforms current approaches.
Nestedness in global trade network revealed using multi-layer approach.
problem Understanding nested structures in international trade networks.
method Constructed a multi-layer network with countries, industries, and transactions; computed nestedness based on buyers' and sellers' involvement.
result Identified variations of nestedness over time and contributing countries and industries.
PNNs model aleatoric uncertainty in scientific machine learning with high accuracy.
problem Aleatoric uncertainty in scientific systems with unequal variance.
method Developed a probabilistic distance metric to optimize PNN architecture and used it in material science applications.
result PNNs yield remarkably accurate output mean estimates and high correlation in predicted intervals.
Poor economies face frequent disruptions that trap them in producing simpler goods.
problem Frequent disruptions in poor economies prevent them from producing complex goods.
method Modeling an evolving input-output network with optimizing agents that adapt to disruptions.
result A poverty trap emerges where disruptions persist despite agents producing simpler goods.
SEF-M improves spiking neural network classification accuracy by 14%.
problem Improving spiking neural network classification accuracy.
method Meta-neuron based learning algorithm with time-varying weight model.
result Time-varying weight model improves classification accuracy by 14%.
Deep sigmoidal networks can achieve dynamical isometry with orthogonal weight initialization, significantly speeding up learning.
problem Ensuring efficient learning in deep neural networks, especially with nonlinear activation functions.
method Employing free probability theory to compute the singular value distribution of a deep network's input-output Jacobian.
result Deep sigmoidal networks can achieve dynamical isometry with orthogonal weight initialization, leading to faster learning.
GENNI visualises neural network symmetries for better identifiability.
problem Identifying functionally equivalent parameters in neural networks.
method Proposes GENNI, an algorithm to efficiently visualise equivalence classes.
result Allows better exploration of identifiability questions in neural networks.
Researchers infer firm-level supply chain networks from sector-level data to assess systemic risk.
problem Estimating systemic risk in economic systems using firm-level data.
method Maximum-entropy algorithms applied to input-output tables and firm-level aggregate output data.
result The most realistic systemic risk content is retrieved by models incorporating disaggregated firm-specific inputs by sector.
Study analyzes catastrophic forgetting in continual learning using teacher-student networks.
problem Catastrophic forgetting in continuously learning systems.
method Teacher-student learning framework, similarity of input distributions and target functions.
result Network can avoid catastrophic forgetting with small input distribution similarity and large target function similarity.
Unified Bayesian framework for LTV system identification using neural networks and Gaussian Processes.
problem Identifying Linear Time-Varying systems from input-output data.
method Bayesian modeling of impulse response as a stochastic process, using neural networks and Gaussian Processes for inference.
result Framework can infer LTI system properties from a single noisy input-output pair, achieving lower error than classical methods.
Survey of methods to verify deep neural networks.
problem Challenges in verifying deep neural networks for specific properties.
method Borrowing insights from reachability analysis, optimization, and search.
result Comparison of existing algorithms and implementations.
Deep learning uses complex networks for high-dimensional data.
problem Computational inefficiency in training deep learning models.
method Use of hierarchical latent variables, efficient linear algebra, SGD optimization, and batch sampling.
result Efficient training and inference possible with optimized algorithms.
Sandpile Economics explains how economies can be prone to large crises from small shocks.
problem Capitalist economies' recurrent crises disproportionate to shocks.
method Formal framework interpreting instability as geometric fragility of production networks.
result Curvature of production networks predicts medium-run output dynamics and resilience.
Different neural networks learn similar mappings with different weights.
problem Understanding shared representations across neural networks with varying weights.
method Shared response model and orthogonal transformations.
result Different neural networks encode the same input examples as different orthogonal transformations of an underlying shared representation.
Exact spectral norm regularization improves neural network generalization.
problem Improving neural network generalization while protecting against noise.
method Exact spectral norm regularization of the Jacobian.
result Improved generalization performance compared to previous methods.