Proposes a feature leveling method to improve interpretability of deep neural networks.
problem Deep neural networks are hard to interpret due to mixed feature levels.
method Introduces a feature leveling architecture to isolate low and high level features.
result Modified models achieve competitive results and improved interpretability.
A ranking network improves neural architecture search efficiency.
problem Efficiently search for neural network architectures without extensive training.
method Pairwise ranking loss for a performance predictor trained on task meta-features.
result The ranking network outperforms the performance predictor in architecture search.
In this paper, we present a novel unsupervised feature learning architecture, which consists of a multi-clustering integration module and a variant of RBM termed multi-clustering integration RBM (MIRBM). In the multi-clustering integration module, we apply three unsupervised K-means, affinity propagation and spectral c…
Automatically finds efficient multi-task models with less data.
problem Training separate models requires more data, parameters, and time.
method Compact search space for multi-task architectures, feature distillation for quick evaluation.
result Automatically identifies multi-task architectures that balance resource requirements and performance.
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 …
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.
Study reveals how neural network architectures bias their learning based on feature directions.
problem Understanding how neural network architectures bias their learning based on feature directions.
method Defined neural anisotropy directions (NADs) to encapsulate the directional inductive bias of architectures and provided an efficient method to identify them.
result NADs characterize the features used by CNNs to discriminate between different classes for the CIFAR-10 dataset.
City-GAN learns city architecture styles using a custom GAN architecture.
problem Learning architectural styles of cities using standard GAN and CGAN architectures.
method Proposes a custom GAN architecture to improve learning of city architectural styles.
result Demonstrates superior performance of custom GAN architecture for city architectural style learning.
AdaEnsemble learns adaptive feature interactions for CTR prediction.
problem Learning feature interactions for CTR prediction in recommender systems and Ads ranking.
method AdaEnsemble is a Sparsely-Gated Mixture-of-Experts (SparseMoE) architecture that dynamically selects feature interaction depth.
result AdaEnsemble achieves better prediction accuracy and inference efficiency compared to state-of-the-art models.
We propose a method to train generative adversarial networks on mutivariate feature vectors representing multiple categorical values. In contrast to the continuous domain, where GAN-based methods have delivered considerable results, GANs struggle to perform equally well on discrete data. We propose and compare several …
Modern audio source separation techniques rely on optimizing sequence model architectures such as, 1D-CNNs, on mixture recordings to generalize well to unseen mixtures. Specifically, recent focus is on time-domain based architectures such as Wave-U-Net which exploit temporal context by extracting multi-scale features. …
This study presents a novel, deep, fully convolutional architecture which is optimized for the task of EEG-based neonatal seizure detection. Architectures of different depths were designed and tested; varying network depth impacts convolutional receptive fields and the corresponding learned feature complexity. Two deep…
Transformers predict price movements from limit order books.
problem Predicting price movements from limit order books.
method Causal convolutional network with masked self-attention.
result Significantly outperforms existing architectures on FI-2010 dataset.
Sensor fusion is a key technology that integrates various sensory inputs to allow for robust decision making in many applications such as autonomous driving and robot control. Deep neural networks have been adopted for sensor fusion in a body of recent studies. Among these, the so-called netgated architecture was propo…
DNF-Net tackles tabular data challenges with neural architecture.
problem Handling tabular data efficiently using neural networks.
method DNF-Net uses a neural architecture with inductive bias corresponding to logical Boolean formulas in disjunctive normal form over affine soft-threshold decision terms.
result DNF-Net significantly outperforms fully connected networks on tabular data.
MixPath unifies multi-path neural architecture search with one-shot training.
problem Efficiently searching multi-path neural architectures.
method One-shot multi-path supernet with Shadow Batch Normalization (SBN).
result SBN stabilizes optimization and improves ranking performance.
DC-NAS improves neural architecture search by clustering and evaluating sub-networks.
problem Inaccurate evaluation of neural architectures in large search spaces.
method Divide-and-Conquer approach: feature representation, clustering, and evaluation of clusters.
result Achieved 75.1% top-1 accuracy on ImageNet, surpassing state-of-the-art methods.
In this paper, we propose Generative Adversarial Network (GAN) architectures that use Capsule Networks for image-synthesis. Based on the principal of positional-equivariance of features, Capsule Network's ability to encode spatial relationships between the features of the image helps it become a more powerful critic in…
Low level features like edges and textures play an important role in accurately localizing instances in neural networks. In this paper, we propose an architecture which improves feature pyramid networks commonly used instance segmentation networks by incorporating low level features in all layers of the pyramid in an o…
EmbraceNet fusion model for multi-sensor activity recognition.
problem Human activity recognition using multiple sensors is challenging.
method Independent processing of each sensor, feature fusion with EmbraceNet, post-processing, and additional processes.
result Improved performance in SHL recognition challenge.
MTL-NAS combines NAS with GP-MTL for task-agnostic multi-task learning.
problem Designing architectures for diverse tasks with varying priors.
method Disentangled GP-MTL networks, hierarchical feature sharing, and gradient-based search.
result General-purpose model trained once can adapt to multiple tasks.
Paper introduces methods to integrate external knowledge into RNNs using attention mechanisms.
problem Incorporating external knowledge into RNNs for improved performance.
method Proposes three methods: attentional concatenation, feature-based gating, and affine transformation.
result Attentional feature-based gating consistently improves performance across tasks.
A new framework learns system design using neural features in function space.
problem Learning system design with neural feature extractors.
method Introduces feature geometry in function space, nesting technique for optimal feature approximation.
result Optimal features found from data samples using off-the-shelf architectures and optimizers.
LOCO-Reg improves CNN accuracy by promoting feature cohesion near filter centers.
problem Current regularization schemes in CNNs violate the principle that weights near the center of a filter are larger than weights on the outside.
method Introduces Locality-Promoting Regularization (LOCO-Reg) to correct this issue.
result LOCO-Reg yields accuracy gains across multiple architectures and datasets.
Study identifies key brain regions and model architectures for ASD diagnosis.
problem Subjective and time-consuming ASD diagnosis by clinicians.
method Comparative analysis of model architectures and atlas granularities.
result High-performing models use 2-4 hidden layers and 16-64 neurons per layer, and cerebellum is predictive of ASD.
xDeepInt learns both vector-wise and bit-wise feature interactions.
problem Learning feature interactions for CTR prediction and recommendation.
method Polynomial Interaction Network (PIN) architecture with subspace-crossing mechanism.
result xDeepInt outperforms state-of-the-art models in CTR prediction and recommendation.
Bayesian optimization (BO) is an effective method of finding the global optima of black-box functions. Recently BO has been applied to neural architecture search and shows better performance than pure evolutionary strategies. All these methods adopt Gaussian processes (GPs) as surrogate function, with the handcraft sim…
Introduces top-k regularization for better feature selection in machine learning.
problem Limited ability of existing feature selection methods to reconcile feature representativeness and inter-correlations.
method Top-k regularization, which induces a sub-architecture on the model's architecture to select informative features and model complex relationships. result Uniform approximation error bound for top-k regularization approximating high-dimensional sparse functions. This paper analyzes the use of 3D Convolutional Neural Networks for brain tumor segmentation in MR images. We address the problem using three different architectures that combine fine and coarse features to obtain the final segmentation. We compare three different networks that use multi-resolution features in terms of…
Speech emotion recognition system using features and text.
problem Improving accuracy in emotion recognition from speech.
method Used speech features (Spectrogram, MFCC) and text, trained Deep Neural Networks.
result Combined MFCC-Text CNN model achieved highest accuracy.
First steps towards a mathematical theory of deep convolutional neural networks for feature extraction were made---for the continuous-time case---in Mallat, 2012, and Wiatowski and Bölcskei, 2015. This paper considers the discrete case, introduces new convolutional neural network architectures, and proposes a mathemati…
Boosting CNNs with dynamic feature selection and boosting weights improves accuracy and efficiency.
problem Expensive training of deep CNNs and manual architecture design.
method Dynamic feature selection, importance sampling, and boosting weights integrated into network training.
result Boosted CNN variants consistently outperform conventional CNNs in accuracy and training speed.
MemNet optimizes neural architectures for memory efficiency.
problem Memory constraints in mobile devices limit the use of large neural networks.
method Augment-trim learning with memory consumption ranking score.
result MemNet finds architectures with 24.17% less memory usage compared to state-of-the-art methods.
SCARLET-NAS improves neural architecture search by stabilizing feature inconsistencies.
problem Difficulty in ranking subnetworks due to feature inconsistency in skip connections.
method Introduced an equivariant learnable stabilizer to homogenize feature disparities.
result SCARLET-A achieves 76.9% top-1 accuracy on ImageNet.
Wind speed prediction improved using a novel deep ensemble learning model inspired by jet aerodynamics.
problem Accurate wind speed forecasting for renewable energy production.
method Proposes a novel Deep Ensemble Learning using Jet-like Architecture (DEL-Jet) to enhance robustness and generalization of a learning system.
result The DEL-Jet technique improves the robustness and generalization of the learning system, as shown by performance evaluations.
The study proves Gaussian universality of deep random features learning.
problem Understanding the test error in deep random features learning.
method Proving Gaussian universality of test error in ridge regression and arbitrary convex losses.
result Sharp asymptotic formula for test error in ridge regression setting.
We consider a neural network architecture with randomized features, a sign-splitter, followed by rectified linear units (ReLU). We prove that our architecture exhibits robustness to the input perturbation: the output feature of the neural network exhibits a Lipschitz continuity in terms of the input perturbation. We fu…
AutoFIS automatically selects important feature interactions for CTR prediction models.
problem Manual feature interaction design is inefficient and prone to noise.
method Two-stage algorithm: search stage relaxes feature interactions to continuous parameters, re-train stage refines model performance.
result AutoFIS significantly improves CTR and CVR of FM-based models.
Approach for selecting features by discarding nuisance and correlated ones.
problem Large datasets with correlated and nuisance features.
method Laplacian score criterion, autoencoder architecture, concrete layer.
result Outperforms similar approaches in clustering performance.
Study evaluates consistency of feature attribution in deep learning for multi-omics data.
problem Challenges in interpretability of deep learning models in biological research.
method Investigation of Shapley Additive Explanations (SHAP) on multi-view deep learning models applied to multi-omics data.
result SHAP rankings are sensitive to architecture and random initialization, suggesting caution.
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.
Deep neural networks and decision trees operate on largely separate paradigms; typically, the former performs representation learning with pre-specified architectures, while the latter is characterised by learning hierarchies over pre-specified features with data-driven architectures. We unite the two via adaptive neur…
Dida learns meta-features invariant to feature permutations.
problem Learning invariant meta-features from dataset samples.
method Distribution-based deep architecture inheriting universal approximation and robustness.
result Dida outperforms state-of-the-art methods on dataset-level tasks.
In recent years, many publications showed that convolutional neural network based features can have a superior performance to engineered features. However, not much effort was taken so far to extract local features efficiently for a whole image. In this paper, we present an approach to compute patch-based local feature…
Deep learning rules simplified with implicit methods.
problem Complexity and interpretability in deep learning models.
method Simplified notation through fixed-point equations and implicit features.
result Simplified deep learning models with enhanced interpretability.
GrateTile optimizes CNN feature map storage for efficient data access.
problem Efficient storage and access of sparse CNN feature maps.
method Divides feature maps into uneven-sized subtensors, compresses and stores them in a compressed yet accessible format.
result Average 55% DRAM bandwidth reduction with minimal indexing overhead.
Improved RA detection with SNN outperforming baseline by 26.8% EER.
problem Improving RA detection systems' generalizability and discriminability.
method Multi-task learning with Siamese Neural Networks (SNN) and additional reconstruction loss.
result SNN outperforms baseline by 26.8% EER, and further improvement by 13.8% with reconstruction loss.
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.