FCC-GAN combines fully connected and convolutional layers for improved GAN performance.
problem Lack of understanding in choosing GAN network architectures.
method Proposes FCC-GAN, a hybrid architecture combining fully connected and convolutional layers.
result FCC-GAN outperforms traditional GAN architectures in terms of learning speed and sample quality.
Hybrid deep architectures with reasoning layers show promising convergence and generalization properties.
problem Understanding the theoretical foundations of hybrid deep architectures with reasoning layers.
method Analyzing the interplay between algorithm layers and neural components in deep architectures.
result Properties of algorithm layers are closely related to the approximation and generalization abilities of end-to-end models.
We characterize convolutional neural networks with respect to the relative amount of features per layer. Using a skew normal distribution as a parametrized framework, we investigate the common assumption of monotonously increasing feature-counts with higher layers of architecture designs. Our evaluation on models with …
Proposes continuous convolution layers for flexible feature map resizing.
problem Fixed stride limitations in discrete convolution layers.
method Introduces Continuous Convolution (CC) layers that use learned continuous functions.
result Dynamic and consistent resizing of feature maps at any scale, non-integer and axis-dependent.
New CNN initialization scheme derived from modern architectures.
problem Stability of CNN model parameters initialization.
method Derived new initialization scheme from modern CNN architectures.
result New initialization method outperforms de facto standard schemes.
The performance of Feedforward neural network (FNN) fully de-pends upon the selection of architecture and training algorithm. FNN architecture can be tweaked using several parameters, such as the number of hidden layers, number of hidden neurons at each hidden layer and number of connections between layers. There may b…
A network architecture improves classification performance on various datasets.
problem Improving classification performance on diverse datasets.
method Divide fully connected layer into three levels, learn existing layers, cluster similar classes, reclassify using clustering masks.
result Achieved state-of-the-art performance with an error rate of 11.56% on Cifar-100.
Graph Metanetworks process diverse neural architectures efficiently.
problem Processing diverse neural architectures efficiently.
method Builds metanetworks using graph neural networks to process graphs representing input neural networks.
result Proves GMNs are expressive and equivariant to parameter permutation symmetries.
Optimizes CNN architectures by analyzing receptive fields without training.
problem CNNs often have too many layers, making them resource-intensive and less effective.
method Layer-wise analysis of receptive fields to identify unproductive layers.
result Identifies and removes unproductive layers, improving CNN performance and efficiency.
Estimates and compares the capabilities of neural network architectures.
problem Developing quantitative methods to estimate and compare neural network architectures.
method Defining capacity as the logarithm of the number of functions an architecture can compute, and proving a cubic polynomial formula for layered architectures.
result The capacity of a neural network is given by a cubic polynomial in the layer sizes, with bottlenecks affecting the capacity.
This paper proposes reusing CNN layers to speed up hyperparameters tuning.
problem Time-consuming hyperparameters tuning in CNNs.
method Reuse trained convolutional layers among different trainings.
result Reduces training time and increases accuracy of neural networks.
DNArch learns CNN architectures by backpropagation.
problem Discovering optimal CNN architectures.
method Differentiable Neural Architectures (DNArch) learns CNN architectures by backpropagation, controlling kernel sizes, channels, downsampling positions, and depth.
result DNArch finds performant CNN architectures across various tasks.
The paper analyzes different neural network architectures for 3D point cloud processing.
problem Understanding the effects of specific intermediate-layer network architectures on 3D point cloud processing.
method Designing five metrics to diagnose various types of DNNs and conducting comparative studies.
result The hypotheses on the effects of specific intermediate-layer network architectures on the representation capacity of DNNs are verified.
New findings show modern neural networks have finite sample complexity in o-minimal structures.
problem Understanding the learnability of modern neural networks in a broad context.
method Analyzing feedforward neural networks definable in o-minimal structures.
result Modern neural networks, including MLPs, CNNs, GNNs, and transformers, have finite sample complexity in the agnostic PAC setting.
Polynomial-time convex optimization for CNNs with ReLU activations.
problem Training Convolutional Neural Networks (CNNs) with ReLU activations.
method Developed a convex analytic framework using semi-infinite duality to formulate equivalent convex optimization problems for CNN architectures.
result Proved that two-layer CNNs can be globally optimized via an ℓ2 norm regularized convex program. This study examines how neural network architecture parameters affect loss surface modality.
problem Understanding the relationship between neural architecture parameters and loss surface modality.
method Fitness landscape analysis of neural network loss surfaces under various architecture settings.
result An increase in problem dimensionality, hidden layer width, and architecture depth affects the modality of loss surfaces.
This paper presents OptNet, a network architecture that integrates optimization problems (here, specifically in the form of quadratic programs) as individual layers in larger end-to-end trainable deep networks. These layers encode constraints and complex dependencies between the hidden states that traditional convoluti…
NeuralArTS categorizes neural ops in a type system for NAS.
problem Manual optimization of search spaces for NAS is inefficient.
method Developed NeuralArTS, a type system for categorizing network ops.
result NeuralArTS can be applied to convolutional layers.
Bayesian method learns neural network architecture parameters.
problem Estimating optimal neural network architecture parameters.
method Bayesian learning of concrete distributions over layer size and network depth.
result Regular networks with learnt structure generalize better on small datasets, while stochastic networks are more robust to initialisation.
A robot learns to classify images with limited perception using a layered reinforcement learning approach.
problem Image classification for robots with partial perception.
method Three-layer architecture using deep reinforcement learning, including meta-layer, action-layer, and classification-layer.
result The method achieves high accuracy on the MNIST dataset and provides explainability of the agent's decision-making process.
Relational Networks (RN) as introduced by Santoro et al. (2017) have demonstrated strong relational reasoning capabilities with a rather shallow architecture. Its single-layer design, however, only considers pairs of information objects, making it unsuitable for problems requiring reasoning across a higher number of fa…
Deep neural-kernel models combine neural networks and kernel machines for scalable large datasets.
problem Combining neural networks and kernel machines for efficient large-scale learning.
method Hybrid neural-kernel architecture using explicit feature mapping and pooling layers.
result The deep neural-kernel models are effective and scalable on benchmark datasets.
A novel multi-layer architecture for one-class classification using graph-embedded kernel ridge regression.
problem Outlier detection in one-class classification using only normal samples.
method Stacking various Graph-Embedded Kernel Ridge Regression (KRR) based Auto-Encoders in a hierarchical fashion.
result The proposed method outperforms existing one-class classifiers on 21 benchmark datasets.
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.
Paper hypothesizes MLP layers in LLMs can be approximated by sparse Mixture of Experts.
problem Understanding dense MLP layers in LLMs.
method Theoretical connection between MoE models and SAE structure in activation space.
result MLP layers in LLMs can be well approximated by sparse Mixture of Experts.
WM layer improves CNN robustness to noise and adversarial attacks.
problem CNNs' susceptibility to noise and adversarial attacks.
method WM layer as a generic architectural addition to CNNs.
result WM variants enhance robustness to noise and adversarial attacks.
Law explains how deep networks separate data for classification.
problem Black-box nature of deep learning limits architecture design and interpretation.
method Studied how deep neural networks process data in intermediate layers.
result Law of geometric data separation emerges in various architectures and datasets.
New neural network architectures use signed permutation representations for finite groups, improving performance.
problem Designing and optimizing neural networks for finite groups with signed permutation representations.
method Introduces G-invariant deep neural networks with densely connected layers and signed permutation representations. result Signed permutation representations lead to significantly better performance in classification tasks.
Paper presents an ADMM-based approach to efficiently integrate quadratic programming layers into neural networks.
problem Integrating quadratic programs into neural networks for optimization.
method An ADMM-based network layer architecture for solving quadratic programs efficiently.
result The ADMM layer is approximately an order of magnitude faster than existing methods for medium scaled problems.
Dual neural network architecture improves accuracy and interpretability.
problem Improving neural network interpretability and accuracy.
method Stacked recurrent and feedforward layers, binary activation function.
result Binary activation leads to simpler, more interpretable models with higher accuracy.
New CNN layer selects important channels to improve model capacity.
problem Improving model capacity under resource constraints.
method Selective allocation of channels in convolutional layers.
result New layer allows new optima that generalize better.
Traditionally, when generative models of data are developed via deep architectures, greedy layer-wise pre-training is employed. In a well-trained model, the lower layer of the architecture models the data distribution conditional upon the hidden variables, while the higher layers model the hidden distribution prior. Bu…
Optimizes crypto-oriented neural architectures for faster secure inference.
problem Privacy conflicts between model users and providers in neural network applications.
method Proposes a novel Partial Activation layer to optimize the initial design of crypto-oriented neural architectures.
result Significant improvement in the efficiency of secure inference on common evaluation metrics.
Paper reviews ranking systems in multi-layered architectures.
problem Ranking systems for effective machine learning models and real-time user responses.
method Examines data processing, representation learning, candidate selection, and online inference layers.
result Ranking systems are complex systems of multiple architectures.
Petridish efficiently searches neural architectures by iteratively adding shortcut connections.
problem Finding efficient neural architectures for various tasks.
method Iteratively adds shortcut connections to existing network layers, motivated by feature selection.
result Petridish efficiently finds competitive models with few GPU days.
We propose a novel family of connectionist models based on kernel machines and consider the problem of learning layer-by-layer a compositional hypothesis class, i.e., a feedforward, multilayer architecture, in a supervised setting. In terms of the models, we present a principled method to "kernelize" (partly or complet…
Paper analyzes infinite-width attention layers using Tensor Programs.
problem Capturing the infinite-width limit of attention layers.
method Tensor Programs framework to rigorously identify the limit distribution.
result Derives exact form of infinite-width limit distribution without Gaussian approximations.
This paper compares Transformers and RNNs in various tasks, showing size differences.
problem Comparing representational capabilities of Transformers and RNNs across tasks.
method Analysis of differences in tasks like index lookup, nearest neighbor, and string equality.
result Size differences in Transformers and RNNs for various tasks.
A new method for verifying deep learning architectures on FPGAs is proposed.
problem Design-time verification of deep learning architectures on FPGAs.
method 2-Level 3-Way (2L-3W) hardware-software co-verification methodology.
result Layer-by-layer similarity scores of 99% accuracy for successful mappings.
Introduces fuzzy layers to enhance deep learning performance.
problem Improving deep learning accuracy for complex tasks.
method Incorporates fuzzy methodologies into deep learning architecture.
result Preliminary results show promising performance in semantic segmentation.
When using deep, multi-layered architectures to build generative models of data, it is difficult to train all layers at once. We propose a layer-wise training procedure admitting a performance guarantee compared to the global optimum. It is based on an optimistic proxy of future performance, the best latent marginal. W…
Transformer is reinterpreted as a numerical ODE solver, leading to improved architectures.
problem Understanding and improving the Transformer architecture.
method Interpreted Transformer as a numerical ODE solver for a multi-particle dynamic system, proposing new architectures.
result Macaron Net outperforms Transformer on supervised and unsupervised learning tasks.
GroSS enables efficient search for grouped convolutional architectures.
problem Training grouped convolutional architectures efficiently and effectively.
method GroSS: Group-Size Series Decomposition for Grouped Architecture Search.
result Simultaneous training of differing numbers of groups within a single layer and all possible combinations between layers.
The study compares feed-forward and attention layers in language models.
problem Understanding the role of feed-forward and attention layers in language models.
method Empirical and theoretical analysis in a synthetic setting.
result Feed-forward layers learn simple distributional associations, while attention layers focus on in-context reasoning.
DNAS disentangles neural architecture search for better interpretability and performance.
problem Lack of interpretability in existing neural architecture search methods.
method DNAS disentangles the hidden representation of the controller into semantically meaningful concepts.
result DNAS achieves state-of-the-art performance and competitive architectures.
New conditions ensure deep neural networks can approximate any function on non-Euclidean spaces.
problem Understanding how to modify neural network architectures to approximate functions on non-Euclidean spaces.
method Developed conditions for feature and readout maps that preserve universal approximation capabilities.
result Modified architectures can deterministically approximate any classifier on non-Euclidean spaces.
This study examines how model architecture affects deep learning model privacy.
problem Privacy concerns in deep learning models due to potential leakage of sensitive information.
method Investigation of CNNs and Transformers, focusing on activation layers, stem layers, LN layers, and attention modules.
result Transformers generally exhibit higher vulnerability to privacy attacks than CNNs.
Understanding the representational power of Restricted Boltzmann Machines (RBMs) with multiple layers is an ill-understood problem and is an area of active research. Motivated from the approach of \emph{Inherent Structure formalism} (Stillinger & Weber, 1982), extensively used in analysing Spin Glasses, we propose a no…