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.

168,932 papers · 148 categories

Trend · papers per month

4897145193 · Jun 202019922001200920172026
48 results for Module Sequencing

New method learns both module structure and sequencing in neural networks.

problem Learning only the parameters and order of execution of neural modules.
method Expands the approach to learn the internal structure of modules, including the ordering and combination of arithmetic operators.
result Performance comparable to hand-designed modules achieved without extra supervisory signals.

Paper proposes a new pipeline for few-shot classification using forget-update module and channel vector sequence.

problem Few-shot classification with limited support samples.
method Channel vector sequence construction module and forget-update module.
result Pipeline achieves state-of-the-art results on various datasets.

We introduce two invariants called sl(3) Khovanov module and pointed sl(3) Khovanov homology for spatial webs (bipartite trivalent graphs). Those invariants are related to Kronheimer-Mrowka's instanton invariants JJ^\sharp and II^\sharp for spatial webs by two spectral sequences. As an application of the spectral seq…

2018-09-13abs ↗pdf ↗

FI-modules were introduced by the first three authors in [CEF] to encode sequences of representations of symmetric groups. Over a field of characteristic 0, finite generation of an FI-module implies representation stability for the corresponding sequence of S_n-representations. In this paper we prove the Noetherian pro…

2012-10-05abs ↗pdf ↗

We study a module structure on Khovanov homology, which we show is natural under the Ozsvath-Szabo spectral sequence to the Floer homology of the branched double cover. As an application, we show that this module structure detects trivial links. A key ingredient of our proof is that the H_1/Torsion module structure on …

2012-04-04abs ↗pdf ↗

Given a Heegaard splitting of a closed 3-manifold, the skein modules of the two handlebodies are modules over the skein algebra of their common boundary surface. The zeroth Hochschild homology of the skein algebra of a surface with coefficients in the tensor product of the skein modules of two handlebodies is interpret…

2004-05-20abs ↗pdf ↗

TFiLM expands convolutional models' receptive field with minimal overhead.

problem Capturing long-range dependencies in sequential data.
method A novel architectural component using a recurrent neural network to modulate convolutional model activations.
result TFiLM significantly improves learning speed and accuracy on various tasks.

New quantum knot invariants derived from Verma modules.

problem Constructing universal quantum knot invariants from Verma modules.
method Defining level N universal invariants from finite quotients of Verma modules over quotient rings.
result Maximal universal invariants for prime N, interpolating Jones and ADO polynomials.

Proposes a novel network-based neighborhood regression for biological systems.

problem Lack of comprehensive analysis on biological modules using both global and local network data.
method Develops a community-wise least square optimization approach to analyze gene modules and their regulatory strength.
result Achieves exact minimax optimality and linear consistency in identifying gene module associations.

Set-Sequence model learns cross-sectional dynamics directly from time series data.

problem Predicting large cross-sections of time series data with latent cross-sectional dynamics.
method A model that learns cross-sectional structure directly, enhancing expressivity and eliminating manual feature engineering.
result Significantly outperforms strong baselines in equity portfolio optimization and loan risk prediction.

ComiRec framework predicts user interests for personalized recommendations.

problem Predicting user interests from sequential behavior data.
method ComiRec framework captures multiple user interests and balances recommendation accuracy and diversity.
result ComiRec achieves significant improvements over state-of-the-art models in sequential recommendation.

New jet functors generalize classical notions in noncommutative geometry.

problem Defining and understanding jet functors in noncommutative settings.
method Constructing and proving properties of jet functors Jd(n)J_d^{(n)}, Jd[n]J_d^{[n]}, and JdnJ_d^n.
result Holonomic jet functor JdnJ_d^n satisfies jet exact sequence under specific conditions.

When analyzing the genome, researchers have discovered that proteins bind to DNA based on certain patterns of the DNA sequence known as "motifs". However, it is difficult to manually construct motifs due to their complexity. Recently, externally learned memory models have proven to be effective methods for reasoning ov…

2017-02-22abs ↗pdf ↗

A sequence of rational functions in a variable qq is qq-holonomic if it satisfies a linear recursion with coefficients polynomials in qq and qnq^n. We prove that the degree of a qq-holonomic sequence is eventually a quadratic quasi-polynomial. Our proof uses differential Galois theory (adapting proofs regarding hol…

2010-05-25abs ↗pdf ↗

In this paper we present the construction of explicit quasi-isomorphisms that compute the cyclic homology and periodic cyclic homology of crossed-product algebras associated with (discrete) group actions. In the first part we deal with algebraic crossed-products associated with group actions on unital algebras over any…

2017-06-27abs ↗pdf ↗

Conformer encoder reverses sequence in time dimension, affecting decoder training.

problem Reversal of sequence in Conformer encoder impacts decoder training.
method Analyzed initial behavior of decoder cross-attention and proposed methods to avoid flipping.
result Self-attention module of Conformer starts dominating, allowing only reversed information to pass.

For a topological space XX, we introduce a criterion for the FI\rm FI module Hi(Confn(X))H^i({\rm Conf}_n(X)) to be finitely generated and give several applications. For instance, if CC is a finite connected CWCW complex, then X=C×R2X = C \times \mathbb{R}^2 satisfies the criterion. Our main tool is a spectral sequence that we der…

2016-12-19abs ↗pdf ↗

DHEN improves CVR prediction for ads with multitask learning and auxiliary loss.

problem Predicting conversion rates in ad-recommendation systems.
method DHEN integrates multiple feature-crossing modules and uses a multitask learning framework, ablation studies, and self-supervised auxiliary loss.
result DHEN achieves state-of-the-art performance in CVR prediction.

Memory-Augmented Recurrent Networks improve dialogue coherence by expanding conversation history storage.

problem Fixed-size vectors limit dialogue coherence; attention mechanisms are computationally expensive.
method Introduce Neural Turing Machines (NTMs) to provide flexible and permanent storage for dialogue history.
result Improved perplexity performance compared to existing baselines.

We generalize our previous work on categorification of Kauffman bracket skein module of surfaces, by extending our homology to tangles in cylinders over surfaces, F x [0,1]. Our homology of 0-tangles and 1-tangles in D^3 coincides (up to normalization) with Khovanov link homology and the reduced Khovanov link homology.…

2004-10-09abs ↗pdf ↗

New exact sequence links cohomology, automorphisms, and extensions of symmetric quandles.

problem Understanding the structure of extensions and automorphisms in symmetric quandles.
method Derived a four-term exact sequence relating 1-cocycles, second cohomology, and automorphisms.
result Obstruction to automorphisms lies in the second cohomology of symmetric quandles.

An important part of the classical theory of real or complex manifolds is the theory of (smooth, real analytic or complex analytic) vector bundles. With any vector bundle over a manifold (M,F) the sheaf of its (smooth, real analytic or complex analytic) sections is associated which is a locally free sheaf of F-modules,…

2011-10-18abs ↗pdf ↗

The C-spectral sequence was introduced by Vinogradov in the late Seventies as a fundamental tool for the study of algebro-geometric properties of jet spaces and differential equations. A spectral sequence arise from the contact filtration of the modules of forms on jet spaces of a fibring (or on a differential equation…

2001-11-13abs ↗pdf ↗

We show, finitely generated rational VICQ\mathsf{VIC}_{\mathbb Q}-modules and SIQ\mathsf{SI}_{\mathbb Q}-modules are uniformly representation stable and all their submodules are finitely generated. We use this to prove two conjectures of Church and Farb, which state that the quotients of the lower central series of the To…

2016-08-23abs ↗pdf ↗

In this paper we apply the theory of finitely generated FI-modules developed by Church, Ellenberg and Farb to certain sequences of rational cohomology groups. Our main examples are the cohomology of the moduli space of n-pointed curves, the cohomology of the pure mapping class group of surfaces and some manifolds of hi…

2012-07-30abs ↗pdf ↗

In this paper, we show the existence of a sequence of invariant differential operators on a particular homogeneous model G/PG/P of a Cartan geometry. The first operator in this sequence can be locally identified with the Dirac operator in kk Clifford variables, D=(D1,...,Dk)D=(D_1,..., D_k), where $D_i=\sum_j e_j\cdot \partial_{i…

2007-09-29abs ↗pdf ↗

NFM improves deep learning by selectively processing hidden states.

problem Processing entire hidden states in each layer limits modularity and reusability.
method Introduces Neural Function Modules (NFM) with attention, sparsity, and feedback.
result Improves results in classification, generalization, generative modeling, and reinforcement learning.

Representation stability is a phenomenon whereby the structure of certain sequences XnX_n of spaces can be seen to stabilize when viewed through the lens of representation theory. In this paper I describe this phenomenon and sketch a framework, the theory of FI-modules, that explains the mechanism behind it.

2014-04-15abs ↗pdf ↗

The colored Jones function of a knot is a sequence of Laurent polynomials. It was shown by TTQ. Le and the author that such sequences are qq-holonomic, that is, they satisfy linear qq-difference equations with coefficients Laurent polynomials in qq and qnq^n. We show from first principles that qq-holonomic sequence…

2003-06-15abs ↗pdf ↗

Improved recurrent neural networks learn long-term dependencies through multi-scale memory.

problem Capturing long-term dependencies in recurrent neural networks.
method Incremental training of a modular RNN architecture with multi-scale hidden states.
result Incremental training and multi-scale memory enhance RNNs' ability to learn long-term dependencies.

In this paper we study the cohomology of (strict) Lie 2-groups. We obtain an explicit Bott-Shulman type map in the case of a Lie 2-group corresponding to the crossed module A1A\to 1. The cohomology of the Lie 2-groups corresponding to the universal crossed modules $G\to \Aut(G)$ and $G\to \Aut^+(G)$ is the abutment of …

2007-12-13abs ↗pdf ↗

An unsupervised anomaly detection method for irregularly sampled time-series data.

problem Anomaly detection in irregularly sampled or missing valued time-series data.
method Uses LSTM networks with time modulation gates to extract temporal features and SVDD for anomaly labeling.
result Significantly outperforms standard approaches on real-life datasets.

Let GG be a group and NN be a normal subgroup of GG. There exists the group extension GG of G/NG/N by NN. For a GG-module AA which NN acts on trivially and a GG-invariant homomorphism on NN to AA, we obtain a central extension of G/NG/N by AA. By using connection cochains, we exhibit the formula of its extens…

2018-03-13abs ↗pdf ↗

Recent breakthroughs in computer vision and natural language processing have spurred interest in challenging multi-modal tasks such as visual question-answering and visual dialogue. For such tasks, one successful approach is to condition image-based convolutional network computation on language via Feature-wise Linear …

2018-08-03abs ↗pdf ↗

Efficient sparse attention reduces self-attention complexity and improves model performance.

problem Quadratic compute and memory requirements of self-attention for long sequences.
method Content-based sparse attention with dynamic routing module.
result Routing Transformer achieves state-of-the-art performance on various benchmarks.

PICLE uses probabilistic models to efficiently evaluate and compose modules for continual learning.

problem Challenging search space of module compositions in continual learning.
method Probabilistic framework to cheaply compute module compositions' fitness.
result First modular CL algorithm to achieve perceptual, few-shot, and latent transfer.