New image classifier uses hierarchical max-pooling with local pooling.
problem Improving image classification accuracy with variable spatial relationships.
method Introduces a hierarchical max-pooling model with additional local pooling for convolutional neural networks.
result Demonstrates improved performance in estimating image features.
We propose a max-pooling based loss function for training Long Short-Term Memory (LSTM) networks for small-footprint keyword spotting (KWS), with low CPU, memory, and latency requirements. The max-pooling loss training can be further guided by initializing with a cross-entropy loss trained network. A posterior smoothin…
Max-pooling architectures are theoretically analyzed and shown to be globally optimized and generalize well.
problem Theoretical understanding and optimization of max-pooling in deep learning architectures.
method Theoretical analysis of a convolutional max-pooling architecture, focusing on a pattern detection problem.
result Max-pooling architectures can be globally optimized and generalize well, even for highly over-parameterized models.
The paper addresses instability in CNNs' first layer by proving max pooling's shift invariance.
problem Instability in CNNs' first layer, leading to sensitivity to small input shifts.
method Establishing conditions for max pooling's shift invariance and deriving a measure of stability.
result Max pooling approximates a nearly shift-invariant complex modulus under certain conditions.
Deep network solves maze path planning without training.
problem Efficient path planning in large mazes with obstacles.
method Max pooling layers without training.
result Solves mazes with over half a billion nodes in short time.
HGNN learns augmented features for deep multi-task learning.
problem Feature learning for deep multi-task learning.
method Hierarchical Graph Neural Network (HGNN) with two levels of graph neural networks.
result Significant performance improvement in classification tasks.
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.
Convolutional neural networks (CNNs) have achieved remarkable performance in many applications, especially in image recognition tasks. As a crucial component of CNNs, sub-sampling plays an important role for efficient training or invariance property, and max-pooling and arithmetic average-pooling are commonly used sub-…
This paper presents a novel approach to exploit the distinctive invariant features in convolutional neural network. The proposed CNN model uses Scale Invariant Feature Transform (SIFT) descriptor instead of the max-pooling layer. Max-pooling layer discards the pose, i.e., translational and rotational relationship betwe…
Deep neural networks have become commonplace in the domain of reinforcement learning, but are often expensive in terms of the number of parameters needed. While compressing deep neural networks has of late assumed great importance to overcome this drawback, little work has been done to address this problem in the conte…
We introduce a novel loss max-pooling concept for handling imbalanced training data distributions, applicable as alternative loss layer in the context of deep neural networks for semantic image segmentation. Most real-world semantic segmentation datasets exhibit long tail distributions with few object categories compri…
Methods from convex optimization are widely used as building blocks for deep learning algorithms. However, the reasons for their empirical success are unclear, since modern convolutional networks (convnets), incorporating rectifier units and max-pooling, are neither smooth nor convex. Standard guarantees therefore do n…
New approach reduces malware detection memory requirements and speeds up training.
problem Efficiently classifying long sequences of malware detection data.
method Developed a new temporal max pooling method and global channel gating design.
result 116x more memory efficient and 25.8x faster training on original dataset.
Debona improves neural network verification by faster and tighter bounds.
problem Proving adversarial robustness of neural networks is computationally hard.
method Decouples upper and lower bounds computation, re-implements Neurify.
result 94% reduction in runtime for proving robustness of max-pooling layers.
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…
Nonlinearity is crucial to the performance of a deep (neural) network (DN). To date there has been little progress understanding the menagerie of available nonlinearities, but recently progress has been made on understanding the rôle played by piecewise affine and convex nonlinearities like the ReLU and absolute value …
Capsule network (CapsNet) was introduced as an enhancement over convolutional neural networks, supplementing the latter's invariance properties with equivariance through pose estimation. CapsNet achieved a very decent performance with a shallow architecture and a significant reduction in parameters count. However, the …
Paper introduces hierarchical softmax for global hierarchical classification tasks.
problem Improving classification accuracy in tasks with class hierarchies.
method Global hierarchical neural networks using hierarchical softmax.
result Hierarchical softmax outperforms regular softmax in multiple datasets.
Empirical evidence suggests that neural networks with ReLU activations generalize better with over-parameterization. However, there is currently no theoretical analysis that explains this observation. In this work, we provide theoretical and empirical evidence that, in certain cases, overparameterized convolutional net…
Low bit-width weights and activations are an effective way of combating the increasing need for both memory and compute power of Deep Neural Networks. In this work, we present a probabilistic training method for Neural Network with both binary weights and activations, called BLRNet. By embracing stochasticity during tr…
This study compares hierarchical and non-hierarchical models for open-domain multi-turn dialog generation.
problem Which kind of models (hierarchical or non-hierarchical) is better for open-domain multi-turn dialog generation?
method Systematically compared nearly all representative hierarchical and non-hierarchical models over the same experimental settings.
result Nearly all hierarchical models are worse than non-hierarchical models in open-domain multi-turn dialog generation, except for HRAN.
We survey agglomerative hierarchical clustering algorithms and discuss efficient implementations that are available in R and other software environments. We look at hierarchical self-organizing maps, and mixture models. We review grid-based clustering, focusing on hierarchical density-based approaches. Finally we descr…
Paper proposes a Renyi entropy-based method for tuning hierarchical topic models.
problem Tuning hierarchical topic models, especially determining the number of topics at each level, is challenging.
method The paper introduces a Renyi entropy-based metric for quality assessment and a practical tuning concept.
result The proposed method can estimate the number of topics for two hierarchical levels in hARTM model.
Neural NMF discovers hierarchical topics in multilayer data.
problem Detecting latent hierarchical structure in multilayer data.
method Recursive application of nonnegative matrix factorization (NMF) in layers with backpropagation optimization.
result Neural NMF outperforms other hierarchical NMF methods in synthetic and real-world datasets.
New kernel connects deep learning to optimization, improving generalization.
problem Understanding why deep neural networks generalize well at over-parameterization.
method Established Neural Optimization Kernel (NOK) linking deep NN to optimization problems.
result New generalization bound for deep structured approximated NOK architecture.
The paper develops a decision support system for hierarchical text classification of conference proceedings.
problem Classifying documents with a fixed hierarchical structure of topics.
method Developed a weighted hierarchical similarity function to calculate topic relevance, using entropy of words to estimate weights.
result The weighted hierarchical similarity function improves ranking accuracy compared to other methods.
Bayesian Hierarchical Invariant Prediction refines ICP for better scalability and prior integration.
problem Improving computational scalability and invariance testing for causal inference.
method Bayesian Hierarchical structure to test invariance under heterogeneous data.
result Demonstrated improved scalability and potential as an alternative to ICP.
Hierarchical causal models help understand cause and effect in nested data.
problem Learning cause and effect from nested hierarchical data.
method Extend structural causal models and causal graphical models with inner plates, develop graphical identification technique and estimation methods.
result Hierarchical data can enable causal identification even when non-hierarchical data cannot.
Posterior regularization enhances Bayesian hierarchical mixture clustering by improving node separation.
problem High nodal variance in BHMC trees, leading to weak separation between nodes at higher levels.
method Employing Posterior Regularization to impose max-margin constraints on nodes at every level.
result Improves cluster separation in BHMC models, enhancing overall model performance.
Two new Hie-TAN and Hie-TAN-Lite algorithms improve TAN for hierarchical feature spaces.
problem Learning dependencies in hierarchical feature spaces.
method Exploits hierarchical parent-child relationships as constraints to learn a dependency tree.
result Hie-TAN-Lite outperforms Hie-TAN and other methods in predictive accuracy.
Introduces hierarchical hyperbolic spaces for non-experts.
problem Understanding hierarchical hyperbolic spaces for non-experts.
method No specific method mentioned; aimed at non-experts.
result Introduces hierarchical hyperbolic spaces for non-experts.
Hierarchical quandles extend diquandles and multi-quandles for link invariants.
problem Constructing invariants for links colored by quandles.
method Hierarchical quandle colourings to construct cocycle invariants.
result Hierarchical quandles provide new link invariants.
Curious hierarchical reinforcement learning improves learning performance.
problem Combining hierarchical abstraction and curiosity-driven exploration in reinforcement learning.
method Developed a method that combines hierarchical reinforcement learning with curiosity.
result Curiosity can more than double learning performance and success rates.
Boxhead dataset tests autoencoder disentanglement in hierarchical data.
problem Evaluate disentanglement in hierarchical data.
method Introduced Boxhead dataset with hierarchically structured factors, evaluated autoencoder models.
result Hierarchical models outperform single-layer VAEs in disentangling factors.
Improved deep hierarchical VAE with diffusion-based VampPrior.
problem Latent variable generative modeling challenges.
method Hierarchical VAE with amortized diffusion-based VampPrior.
result Better performance with fewer parameters and improved stability.
Study on detecting hierarchical community structures in networks.
problem Detecting hierarchical community structures in networks.
method Analysis of planted hierarchies of partitions in networks, identifying additional detectability phases.
result There are additional phases in which the presence of multiple consistent partitions can either help or hinder detection of hierarchical structures.
Reduces high granularity and dimensionality in hierarchical categorical variables.
problem Overfitting and estimation issues in predictive models due to high granularity and dimensionality.
method Entity embedding and top-down clustering algorithm to reduce granularity and dimensionality.
result The reduced hierarchy improves model fit and complexity balance.
Proposes a hierarchical clustering method for positive and negative dissimilarities.
problem Clustering dissimilarities, especially positive and negative.
method Hierarchical correlation clustering followed by tree preserving embedding.
result Performance on various datasets.
Paper proposes a new method for selecting the best hierarchical forecasting approach.
problem Selecting the best method for reconciling base forecasts in hierarchical time series.
method Conditional hierarchical forecasting using machine learning and time series features.
result Conditional hierarchical forecasting leads to significantly more accurate forecasts, especially at lower levels.
The cooperative hierarchical structure is a common and significant data structure observed in, or adopted by, many research areas, such as: text mining (author-paper-word) and multi-label classification (label-instance-feature). Renowned Bayesian approaches for cooperative hierarchical structure modeling are mostly bas…
A new method for efficient portfolio optimization using graph structures.
problem Optimizing portfolio weights while reducing computational complexity.
method Hierarchical graph structures and Schur complement method.
result Optimal portfolio weights can be computed efficiently by inverting small submatrices.
Survey of tools for studying hierarchical hyperbolic spaces.
problem Understanding hierarchical hyperbolic spaces.
method Various tools developed for studying HHSs.
result Ease of use for non-experts in HHS machinery.
Deep neural networks continue to show improved performance with increasing depth, an encouraging trend that implies an explosion in the possible permutations of network architectures and hyperparameters for which there is little intuitive guidance. To address this increasing complexity, we propose Evolutionary DEep Net…
Learning invariant representations is an important problem in machine learning and pattern recognition. In this paper, we present a novel framework of transformation-invariant feature learning by incorporating linear transformations into the feature learning algorithms. For example, we present the transformation-invari…
This work proposes optimal decision rules for hierarchical classifiers to better align with evaluation metrics.
problem Heuristic decision rules in hierarchical classification do not align with evaluation metrics.
method Derives optimal decision rules for various prediction settings, focusing on hierarchical hFβ scores. result Optimal decision rules enhance the performance and reliability of hierarchical classifiers.
The paper extends fairness to hierarchical clustering, finding efficient algorithms with minimal loss.
problem Ensuring fairness in hierarchical clustering where data is recursively partitioned.
method Extending fairness to hierarchical clustering, developing simple, efficient algorithms for various objectives.
result Simple, efficient algorithms for fair hierarchical clustering with only a negligible loss in objective.
Multicurve stabilizers' extensions are hierarchically hyperbolic.
problem Characterizing the structure of multicurve stabilizers' extensions.
method Proving the extensions of multicurve stabilizers are hierarchically hyperbolic groups.
result Extensions of multicurve stabilizers are hierarchically hyperbolic.
Novel graph network learns hierarchical network structure.
problem Lack of information in hierarchical network topology.
method Hierarchical clustering for multiscale decomposition, graph convolutional layers.
result Competitive performance on citation network benchmark.