Deep ResNets favor low bottleneck rank with proper hyperparameters.
problem Understanding the inductive bias of deep neural networks.
method Computed minimum-norm weights of a deep linear ResNet.
result Deep nonlinear ResNets have an inductive bias towards minimizing bottleneck rank.
Large neural networks learn low-dimensional representations that balance complexity and regularity.
problem Understanding the tradeoff between low-dimensional representations and complexity in deep neural networks.
method Computed finite depth corrections to reveal a measure of regularity that bounds the pseudo-determinant of the Jacobian.
result Proved the conjectured bottleneck structure in learned features as network depth increases, showing almost all hidden representations are approximately low-dimensional and weight matrices have singular values close to 1.
CNNs use a bottleneck structure to focus on a few frequencies, affecting function representation.
problem Understanding how CNNs focus on specific frequencies in their feature learning.
method Defined Convolution Bottleneck (CBN) structure, measured CBN rank, and analyzed parameter norms.
result Parameter norm scales with depth and CBN rank, and networks with optimal parameters exhibit this structure.
PLUMAGE improves large model training efficiency and stability.
problem Accelerator memory and networking constraints during large model training.
method Probabilistic Low rank Unbiased Minimum Variance Gradient Estimator (PLUMAGE) that resolves bias and variance issues.
result PLUMAGE reduces training loss by 28% on average across the GLUE benchmark.
We propose to solve a label ranking problem as a structured output regression task. We adopt a least square surrogate loss approach that solves a supervised learning problem in two steps: the regression step in a well-chosen feature space and the pre-image step. We use specific feature maps/embeddings for ranking data,…
New approach reduces model size for Transformer architectures.
problem Large embedding dimensions limit model applicability.
method Identified low-rank bottleneck in multi-head attention.
result Reducing head size to sequence length improves model performance.
This work analyzes how bottleneck layers and skip connections affect linear denoising autoencoders' generalization.
problem Understanding the generalization of linear denoising autoencoders in overparameterized regimes.
method Analyzes two-layer linear denoising autoencoders with a bottleneck layer and skip connection, deriving test risk formulas.
result Bottleneck layers introduce an additional complexity measure, while skip connections can mitigate variance.
We consider the problem of learning over non-stationary ranking streams. The rankings can be interpreted as the preferences of a population and the non-stationarity means that the distribution of preferences changes over time. Our goal is to learn, in an online manner, the current distribution of rankings. The bottlene…
The Softmax function on top of a final linear layer is the de facto method to output probability distributions in neural networks. In many applications such as language models or text generation, this model has to produce distributions over large output vocabularies. Recently, this has been shown to have limited repres…
We accelerate the power method for strong low-rank approximation using fast sketching.
problem Efficiency bottleneck in power method for large target ranks.
method Developed an algorithmic and theoretical framework for accelerating the power method using fast sketching.
result Simple and provably efficient methods for singular value decomposition, low-rank factorization, and Nyström approximation.
Deep ReLU networks escape from the origin via saddle points with a low-rank bias.
problem Understanding the dynamics of gradient descent in deep ReLU networks.
method Analysis of escape directions and singular values of weight matrices.
result The first singular value of the ℓ-th layer weight matrix is at least ℓ41 larger than any other singular value. cuRegOT accelerates GPU-based entropic OT solving.
problem Slow convergence and high computational cost of optimal transport on GPUs.
method High-performance GPU solver with algorithmic and architectural optimizations.
result Significant speedups over state-of-the-art solvers.
Proposes a faster Isomap algorithm by reducing eigenvalue decomposition complexity.
problem High computational complexity of Isomap, especially in eigenvalue decomposition stage.
method Introduces a projection operator to reduce the complexity of the eigenvalue decomposition stage to linear order.
result Reduces Isomap's computational complexity to linear order while preserving structural information.
RFX accelerates and compresses Random Forests for large datasets.
problem Memory bottleneck in proximity matrices limits Random Forest analysis.
method QLORA compression, CPU TriBlock storage, GPU batch sizing, 3D MDS visualization.
result Proximity-based Random Forest analysis on larger datasets is feasible.
In this paper, we propose a convergent parallel best-response algorithm with the exact line search for the nondifferentiable nonconvex sparsity-regularized rank minimization problem. On the one hand, it exhibits a faster convergence than subgradient algorithms and block coordinate descent algorithms. On the other hand,…
PowerGossip compresses model differences for decentralized deep learning with low-rank linear compressors.
problem Communication bottleneck in decentralized deep learning models.
method Low-rank linear compressors applied on model differences using power iteration steps.
result Asymptotically independent of network and compression, faster convergence, and comparable performance to tuned compression algorithms.
We propose a general framework for reduced-rank modeling of matrix-valued data. By applying a generalized nuclear norm penalty we can directly model low-dimensional latent variables associated with rows and columns. Our framework flexibly incorporates row and column features, smoothing kernels, and other sources of sid…
We study gradient compression methods to alleviate the communication bottleneck in data-parallel distributed optimization. Despite the significant attention received, current compression schemes either do not scale well or fail to achieve the target test accuracy. We propose a new low-rank gradient compressor based on …
New research shows non-bottlenecked autoencoders can outperform bottlenecked ones for anomaly detection.
problem The necessity of a bottleneck in autoencoders for anomaly detection.
method Investigated two ways to remove bottlenecks: overparameterising the latent layer and introducing skip connections. Carried out extensive experiments on various AE types and datasets.
result Non-bottlenecked autoencoders can outperform bottlenecked ones, improving anomaly detection performance.
Develops PRPCA for smooth image recovery combining low-rank and smoothness.
problem Image matrix recovery under low-rank and smoothness assumptions.
method Projected Robust PCA (PRPCA) framework combining low-rank and smoothness.
result Explicit statistical guarantees for PRPCA, reducing matrix dimensionality.
SCBMs model causal effects using low-dimensional bottlenecks.
problem Causal effect estimation in high-dimensional systems.
method Structural causal models with low-dimensional summary statistics.
result SCBMs provide a flexible framework for task-specific dimension reduction.
Wide neural networks with narrow bottlenecks behave like deep Gaussian processes.
problem Understanding the behavior of neural networks with narrow layers in the wide limit.
method Analyzing the wide limit of BNNs with narrow bottlenecks, showing they behave like a composition of GPs.
result Wide neural networks with narrow bottlenecks form a composition of GPs, termed a bottleneck NNGP.
Network node embedding is an active research subfield of complex network analysis. This paper contributes a novel approach to learning network node embeddings and direct node classification using a node ranking scheme coupled with an autoencoder-based neural network architecture. The main advantages of the proposed Dee…
Softmax is an output activation function for modeling categorical probability distributions in many applications of deep learning. However, a recent study revealed that softmax can be a bottleneck of representational capacity of neural networks in language modeling (the softmax bottleneck). In this paper, we propose an…
A self-supervised debiasing method using rank regularization mitigates spurious correlations in neural networks.
problem Spurious correlations cause biases in deep neural networks, affecting generalization.
method Spectral analysis of latent representations, rank regularization, self-supervised pretraining, debiasing of downstream tasks.
result The proposed framework significantly improves generalization performance and outperforms supervised debiasing approaches.
Study shows bottlenecks improve image segmentation quality.
problem Robust object discovery in real-world images remains challenging.
method Empirical investigation of reconstruction bottlenecks in GENESIS model.
result Reconstruction bottlenecks determine reconstruction and segmentation quality.
Paper revisits Deep Variational Information Bottleneck and proposes a new optimization approach.
problem Limitations of Deep Variational Information Bottleneck in optimizing mutual information.
method Proposes a new optimization approach by circumventing the limitation of requiring both Markov chains during optimisation.
result Shows how to optimise a lower bound for mutual information, circumventing the limitation of requiring both Markov chains.
A new method for efficient neural network fine-tuning using queryable low-rank update atoms.
problem Rigidity of static low-rank adaptation methods when input and depth-wise computation vary.
method A shared queryable memory of low-rank update atoms, allowing dynamic and context-sensitive adaptation.
result Improves final test performance and training stability compared to standard low-rank adaptation.
CB-APM uses analyst consensus as a bottleneck to interpret stock returns.
problem Tackles the challenge of understanding and predicting stock returns using professional beliefs.
method Embeds analyst consensus as a structural bottleneck, treating it as a sufficient statistic for market information.
result CB-APM portfolios exhibit strong monotonic return gradients and robust across different economic conditions.
We consider the task of building compact deep learning pipelines suitable for deployment on storage and power constrained mobile devices. We propose a unified framework to learn a broad family of structured parameter matrices that are characterized by the notion of low displacement rank. Our structured transforms admit…
In this paper we propose new techniques to sample arbitrary third-order tensors, with an objective of speeding up tensor algorithms that have recently gained popularity in machine learning. Our main contribution is a new way to select, in a biased random way, only O(n1.5/ε2) of the possible n3 elements while s…
Simplifies transfer learning with deep neural networks using ridge regression.
problem High computational cost of finetuning deep models for transfer learning.
method Leverage the low-rank property of deep neural networks' feature vectors in kernel ridge regression.
result Successful on supervised and semi-supervised transfer learning tasks.
Wide neural networks become linear, but adding bottlenecks makes them bilinear or multilinear.
problem Understanding the transition of neural networks from linearity to higher-order functions.
method Analyzing the behavior of randomly initialized wide neural networks with and without bottleneck layers.
result Bottleneck layers transform the network's function from linear to bilinear or multilinear.
New method quantifies redundant information using information bottleneck.
problem Quantifying redundant information among multiple sources.
method Formulated as an information bottleneck problem, termed redundancy bottleneck.
result Extracts information that best predicts the target without revealing source identity.
Decentralized framework for spatial data inference over vulnerabilities.
problem Vulnerabilities in centralized frameworks for spatial data.
method Low-rank models, decentralized optimization, evidence lower bound, block descent method.
result Consistency and asymptotic normality of the estimator in spatial low-rank models.
Soft-Radial Projection solves gradient saturation in constrained deep learning.
problem Gradient saturation in deep learning models when integrating hard constraints.
method Introduces Soft-Radial Projection, a differentiable layer that maps predictions onto constraint boundaries without rank-deficient Jacobians.
result Improves convergence and solution quality over state-of-the-art methods.
The Mahler volume of a centrally symmetric convex body K is defined as M(K)= (Vol K)(Vol K^dual). Mahler conjectured that this volume is minimized when K is a cube. We introduce the bottleneck conjecture, which stipulates that a certain convex body K^diamond subset K X K^dual has least volume when K is an ellipsoid. If…
The Information bottleneck method is an unsupervised non-parametric data organization technique. Given a joint distribution P(A,B), this method constructs a new variable T that extracts partitions, or clusters, over the values of A that are informative about B. The information bottleneck has already been applied to doc…
MO2 learns useful behaviours from past experience for new tasks.
problem Discovering useful behaviours from past experience and transferring them to new tasks.
method Model-Based Offline Options (MO2) framework supporting sample-efficient bottleneck option discovery over continuous state-action spaces.
result MO2 outperforms recent option learning methods on complex long-horizon continuous control tasks.
Graph neural networks struggle to propagate long-range information, causing over-squashing.
problem Graph neural networks struggle to propagate long-range information.
method Identified over-squashing as the bottleneck in GNNs, demonstrated on various models.
result Breaking the bottleneck improves GNNs' performance on long-range problems.
In this paper, we provide an information-theoretic interpretation of the Vector Quantized-Variational Autoencoder (VQ-VAE). We show that the loss function of the original VQ-VAE can be derived from the variational deterministic information bottleneck (VDIB) principle. On the other hand, the VQ-VAE trained by the Expect…
MPNNs struggle with class-bottlenecks and heterophily, leading to performance limitations.
problem Performance limitations of MPNNs under heterophily and structural bottlenecks.
method A statistical framework decomposing model performance into SNR components and proving bounds on sensitivity.
result Optimal graph structures for maximizing higher-order homophily are disjoint unions of single-class and two-class-bipartite clusters.
Tutorial on information bottleneck problems with connections to coding and learning.
problem Information bottleneck problems and their connections to coding and learning.
method Information theoretic perspective, practical methods, connections to various problems.
result Optimal trade-offs between relevance and complexity in discrete and vector Gaussian frameworks.
CEB enhances model resilience through simple entropy bottleneck.
problem Improving model robustness against adversarial attacks.
method Conditional Entropy Bottleneck (CEB) combined with data augmentation.
result CEB significantly boosts adversarial robustness on various benchmarks.
Deep reinforcement learning has recently shown many impressive successes. However, one major obstacle towards applying such methods to real-world problems is their lack of data-efficiency. To this end, we propose the Bottleneck Simulator: a model-based reinforcement learning method which combines a learned, factorized …
Proposes Decodable Information Bottleneck for optimal representation learning.
problem Finding optimal representations for supervised learning.
method Integrates information retention and compression with the desired predictive family.
result Optimal representations lead to better expected test performance and can be estimated with guarantees.
We introduce the HSIC (Hilbert-Schmidt independence criterion) bottleneck for training deep neural networks. The HSIC bottleneck is an alternative to the conventional cross-entropy loss and backpropagation that has a number of distinct advantages. It mitigates exploding and vanishing gradients, resulting in the ability…
Efficiently samples complex distributions using tensor train format.
problem Sampling from high-dimensional complex probability densities efficiently.
method Integrates tensor train format with backward stochastic differential equations (BSDEs) for fast, robust, and accurate sampling.
result Improved efficiency in sampling from challenging target distributions.