Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,657 papers · 148 categories

Trend · papers per month

3537051,0581,410 · Jun 202019922001200920172026
48 results for Tiny Recursive Models

ThriftyNet uses a single convolutional layer recursively to maximize parameter usage.

problem Maximizing the use of parameters in deep convolutional neural networks.
method A single convolutional layer is used recursively, with normalization, non-linearities, downsampling, and shortcuts to maintain model expressivity.
result ThriftyNet achieves competitive performance with significantly fewer parameters.

LASER compresses recursive model activations by exploiting their low-dimensional structure.

problem Understanding and optimizing the geometric structure of recursive reasoning trajectories.
method Dynamic low-rank basis tracking via matrix-free subspace tracking with a fidelity-triggered reset mechanism.
result Recursive activations occupy a linear, low-dimensional subspace that can be compressed efficiently.

Channel Pruning, widely used for accelerating Convolutional Neural Networks, is an NP-hard problem due to the inter-layer dependency of channel redundancy. Existing methods generally ignored the above dependency for computation simplicity. To solve the problem, under the Bayesian framework, we here propose a layer-wise…

2018-12-02abs ↗pdf ↗

Finding Tiny Faces (by Hu and Ramanan) proposes a novel approach to find small objects in an image. Our contribution consists in deeply understanding the choices of the paper together with applying and extending a similar method to a real world subject which is the counting of people in a public demonstration.

2018-01-19abs ↗pdf ↗

The paper explores generalizations of Mirzakhani's recursion and computes volumes for physical gravity models.

problem Computing volumes for physical gravity models.
method Topological recursion and physical two-dimensional gravity models.
result Derivation of Virasoro constraints and cut-and-join equations for generalized Mirzakhani's recursions.

New algorithm PRACTISE accelerates networks with tiny sets, reducing latency by 22%.

problem Accelerating networks with tiny training sets to improve performance and privacy.
method Proposes a new algorithm PRACTISE based on finetune convexity hypothesis and feature mimicking.
result PRACTISE outperforms previous methods by reducing latency by 22% on ImageNet-1k.

Improved deep learning model deployment on tiny MCUs with mixed-precision quantization.

problem Memory limitations prevent accurate deployment of DNN models on tiny MCUs.
method Automated mixed-precision quantization using Reinforcement Learning for MCU constraints.
result Mixed-precision models achieve high accuracy with uniform quantization policies.

This work analyzes how frequency components affect CNN predictions and robustness.

problem Lack of frequency-based explanation for CNNs leading to vulnerabilities.
method Frequency component analysis and quantification of their contribution to CNN predictions.
result Adversarial attacks exploit high-frequency features, while robustness comes from low-frequency associations.

Study Lie algebras with complex structures, focusing on degenerations and deformations.

problem Understanding the space of Lie algebras with complex structures and their transformations.
method Identifying invariants that remain consistent under degenerations and applying to four-dimensional case.
result Found invariants that help in understanding the behavior of Lie algebras under complex structures.

Recursive training of generative models can lead to model collapse, and the recursion converges to a unique limiting distribution.

problem Model collapse in recursive training of generative models
method Recursive training on their own outputs
result Recursive training converges to a unique limiting distribution

In this article, we aim at improving the prediction of expert aggregation by using the underlying properties of the models that provide expert predictions. We restrict ourselves to the case where expert predictions come from Kalman recursions, fitting state-space models. By using exponential weights, we construct diffe…

2020-02-26abs ↗pdf ↗

The paper studies risk-sensitive MDPs with recursive risk measures.

problem Risk-sensitive decision-making in MDPs with unbounded costs.
method Recursive application of static risk measures, Bellman equation derivation, existence of optimal policies.
result Existence of Markovian optimal policies for infinite planning horizons, contractive model for stationary optimal policy.

This paper studies recursive ensembles driven by Fibonacci updates, improving learning dynamics.

problem Improving learning dynamics in recursive ensemble learning.
method Develops second-order recursive architectures with Fibonacci-type update flows.
result Establishes global convergence conditions and generalization bounds for recursive ensembles.

Theoretical study on AI models' resilience to data contamination during recursive training.

problem Data contamination in recursive training of generative AI models.
method General framework with minimal assumptions on real data distribution and flexible generative models.
result Contaminated recursive training converges with a rate equal to the minimum of baseline model's rate and contamination fraction.

Recurrent neural networks (RNNs) process input text sequentially and model the conditional transition between word tokens. In contrast, the advantages of recursive networks include that they explicitly model the compositionality and the recursive structure of natural language. However, the current recursive architectur…

2016-07-15abs ↗pdf ↗

POET enables large neural network training on tiny devices with reduced energy.

problem Training large neural networks on memory-limited edge devices.
method Jointly optimizes rematerialization and paging for memory reduction, formulating an MILP for energy-efficient training.
result POET trains ResNet-18 and BERT within Cortex-M memory constraints, outperforming current methods in energy efficiency.

A new metric HSW derived from hierarchical Radon Transform addresses computational bottlenecks in sliced Wasserstein.

problem Computational inefficiency of sliced Wasserstein in high-dimensional settings with few supports.
method Hierarchical Radon Transform (HRT) and bottleneck projections to reduce projection number.
result HSW metric derived from HRT is computationally efficient and maintains metric properties.

Study uses reinforcement learning to optimize portfolios under recursive utility.

problem Improving portfolio allocation using risk-sensitive objectives.
method Approximated certainty equivalent via Monte Carlo, trained actor-critic algorithms (PPO, A2C).
result Recursive-utility agent outperforms discounted baseline in Sharpe ratio, max drawdown, and cumulative return.

A new method optimizes diffusion models with recursive likelihood ratios.

problem Efficiently aligning pre-trained diffusion models for specific applications.
method Recursive Likelihood Ratio (RLR) optimizer for Half-Order (HO) fine-tuning.
result The RLR method achieves unbiased and lower-variance gradients, improving model performance.

AL0\ell_0CORE tensor decomposition reduces computational cost for sparse count data.

problem Efficiently decompose sparse count data matrices.
method Probabilistic Tucker decomposition with 0\ell_0-norm constraint.
result AL0\ell_0CORE achieves similar results to full Tucker decomposition at a fraction of the cost.

Harer and Zagier proved a recursion to enumerate gluings of a 2d2d-gon that result in an orientable genus gg surface, in their work on Euler characteristics of moduli spaces of curves. Analogous results have been discovered for other enumerative problems, so it is natural to pose the following question: how large is t…

2018-12-31abs ↗pdf ↗

Improved robustness of machine learning models with controlled Lipschitz constants.

problem Vulnerability of state-of-the-art models to adversarial attacks.
method Proposes a CLL loss that calibrates the margin and Lipschitz constant penalties, improving robustness certificates.
result Consistently outperforms other losses on CIFAR-10, CIFAR-100, and Tiny-ImageNet datasets.

Bayesian method for multivariate autoregressive models with exogenous inputs.

problem Estimating uncertainties in autoregressive models with exogenous inputs.
method Recursive Bayesian estimation via message passing in a factor graph.
result Produces full posterior distributions for autoregressive coefficients and noise precision.

The paper uses tensor decompositions to improve neural network models for tree data.

problem Encoding structural knowledge from tree-structured data efficiently.
method Introduces new aggregation functions using Canonical and Tensor-Train decompositions.
result Proposed models outperform traditional methods on tree classification tasks.

CEFOL uses deep learning for dynamic programming with recursive utility.

problem Challenges in solving dynamic programming problems with recursive utility.
method Introduces a separate neural network for certainty equivalent, uses first-order optimality conditions to learn value and policy functions.
result CEFOL achieves high accuracy in learning value and policy functions, matching VFI benchmarks.

Study adds investment gains and losses to recursive utility model, proving existence and uniqueness of utility process.

problem Existence and uniqueness of utility process in a recursive utility model with investment gains and losses.
method Generalized recursive utility model with constant elasticity of intertemporal substitution and relative risk aversion degree. Proved existence and uniqueness in a specific, finite-state Markovian setting.
result Utility process exists and is unique when agent derives nonnegative gain-loss utility, and non-existent or non-unique otherwise.

This text investigates relations between two well-known family of algorithms, matrix factorisations and recursive linear filters, by describing a probabilistic model in which approximate inference corresponds to a matrix factorisation algorithm. Using the probabilistic model, we derive a matrix factorisation algorithm …

2015-09-07abs ↗pdf ↗

We extend topological recursion to twisted Higgs bundles with singularities.

problem Computing Taylor expansions of period matrices for twisted Higgs bundles.
method We introduce a twisted topological recursion on the spectral curve of a twisted Higgs bundle, encoding singularities and performing the recursion explicitly.
result The g=0g=0 twisted Eynard-Orantin differentials compute the Taylor expansion of the spectral curve's period matrix, independent of the ambient space.

Greedy training of recursive partitioning estimators faces a computational barrier when the true function doesn't satisfy a specific property.

problem Computational inefficiency of greedy training for recursive partitioning estimators.
method Analysis of greedy training for sparse regression functions over binary features.
result Greedy training requires exponential samples when the true function doesn't satisfy a specific property (MSP), but only logarithmic samples when it does.