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.
BCRL learns a Bellman complete representation for offline RL policy evaluation.
HEAR benchmark evaluates audio representations for diverse tasks.
Learning knowledge representation is an increasingly important technology applicable in many domain-specific machine learning problems. We discuss the effectiveness of traditional Link Prediction or Knowledge Graph Completion evaluation protocol when embedding knowledge representation for categorised multi-relational d…
State representation learning aims at learning compact representations from raw observations in robotics and control applications. Approaches used for this objective are auto-encoders, learning forward models, inverse dynamics or learning using generic priors on the state characteristics. However, the diversity in appl…
New methods evaluate data representations by complexity of low-loss predictor learning.
This paper evaluates various representations for robotics tasks, improving performance in lifting, stacking, and pushing.
A new method evaluates invariant performance of IRM-based representations.
The topology of -representation varieties of the fundamental groups of planar webs so that the meridians are sent to matrices with trace equal to are explored, and compared to data coming from spider evaluation of the webs. Corresponding to an evaluation of a web as a spider is a rooted tree. We associate t…
State representation learning, or the ability to capture latent generative factors of an environment, is crucial for building intelligent agents that can perform a wide variety of tasks. Learning such representations without supervision from rewards is a challenging open problem. We introduce a method that learns state…
This paper starts a systematic description of colored knot polynomials, beginning from the first non-(anti)symmetric representation R=[2,1]. The project involves several steps: (i) parametrization of big families of knots a la arXiv:1506.00339, (ii) evaluating Racah/mixing matrices for various numbers of strands in var…
New measures quantify diversity of latent representations using metric space magnitude.
This paper critically examines unsupervised disentangled representation learning, revealing challenges and limitations.
Introduction 1. The two-eigenvalue problem 2. Hecke algebra representations of braid groups 3. Duality of Jones-Wenzl representations 4. Closed images of Jones-Wenzl sectors 5. Distribution of evaluations of Jones polynomials 6. Fibonacci representations
New insights into state representations in RL help design better learning rules.
Deep RL approach improves MIS for complex environments.
We release the largest public ECG dataset of continuous raw signals for representation learning containing 11 thousand patients and 2 billion labelled beats. Our goal is to enable semi-supervised ECG models to be made as well as to discover unknown subtypes of arrhythmia and anomalous ECG signal events. To this end, we…
Paper defines and solves a problem in representation learning to ensure fairness with high confidence.
Obtaining colored HOMFLY-PT polynomials for knots from 3-strand braid carrying arbitrary representation is still tedious. For a class of rank symmetric representations, -colored HOMFLY-PT evaluation becomes simpler. Recently it was shown that , for such knots from 3-strand braid, can…
This paper evaluates fairness in deep metric learning and proposes a method to reduce subgroup performance gaps.
Study reveals limitations of fair representation learning methods and cautions against their use in performance-sensitive tasks.
New method uses contrastively trained GNNs for more reliable graph model evaluation.
With the rising interest in graph representation learning, a variety of approaches have been proposed to effectively capture a graph's properties. While these approaches have improved performance in graph machine learning tasks compared to traditional graph techniques, they are still perceived as techniques with limite…
Derives integral representations for a Lévy process and its extremum, hitting time, with fast evaluation.
Replicated and validated Rank-N-Contrast for robust regression.
Study evaluates scalability and real-world impact of disentangled representations.
A commonly used evaluation metric for text-to-image synthesis is the Inception score (IS) \cite{inceptionscore}, which has been shown to be a quality metric that correlates well with human judgment. However, IS does not reveal properties of the generated images indicating the ability of a text-to-image synthesis method…
LCIT tests conditional independence using latent representations.
DECAT framework evaluates multimodal models for shared biology, detecting confounders and false positives.
Variational autoencoders (VAEs) learn representations of data by jointly training a probabilistic encoder and decoder network. Typically these models encode all features of the data into a single variable. Here we are interested in learning disentangled representations that encode distinct aspects of the data into sepa…
Learning disentangled representation from any unlabelled data is a non-trivial problem. In this paper we propose Information Maximising Autoencoder (InfoAE) where the encoder learns powerful disentangled representation through maximizing the mutual information between the representation and given information in an unsu…
The study uncovers latent capabilities of language models via causal representation learning.
Self-supervised methods, wherein an agent learns representations solely by observing the results of its actions, become crucial in environments which do not provide a dense reward signal or have labels. In most cases, such methods are used for pretraining or auxiliary tasks for "downstream" tasks, such as control, expl…
Deep-embedding methods aim to discover representations of a domain that make explicit the domain's class structure and thereby support few-shot learning. Disentangling methods aim to make explicit compositional or factorial structure. We combine these two active but independent lines of research and propose a new parad…
Bayesian algorithm improves word representations using semantic taxonomy.
We make two theoretical contributions to disentanglement learning by (a) defining precise semantics of disentangled representations, and (b) establishing robust metrics for evaluation. First, we characterize the concept "disentangled representations" used in supervised and unsupervised methods along three dimensions-in…
We analyze disentangled representations under a causal generative process, proposing new metrics and datasets.
Self-supervised learning improves ECG classification performance.
Pix2Shape learns 3D scene representations from single images without supervision.
URL benchmark evaluates uncertainty quantification in pretrained models.
Recently several researchers have investigated techniques for using data to learn Bayesian networks containing compact representations for the conditional probability distributions (CPDs) stored at each node. The majority of this work has concentrated on using decision-tree representations for the CPDs. In addition, re…
Researchers propose a new SSL risk decomposition method to evaluate and improve self-supervised learning models.
Dual representations for robust risk measures and uncertainty sets.
Proposes GM Score to evaluate GANs considering diversity, disentanglement, and discriminability.
Learning knowledge representation is an increasingly important technology that supports a variety of machine learning related applications. However, the choice of hyperparameters is seldom justified and usually relies on exhaustive search. Understanding the effect of hyperparameter combinations on embedding quality is …
This paper presents a novel clustering concept that is based on jointly learned nonlinear transforms (NTs) with priors on the information loss and the discrimination. We introduce a clustering principle that is based on evaluation of a parametric min-max measure for the discriminative prior. The decomposition of the pr…
We study the problem of off-policy policy evaluation (OPPE) in RL. In contrast to prior work, we consider how to estimate both the individual policy value and average policy value accurately. We draw inspiration from recent work in causal reasoning, and propose a new finite sample generalization error bound for value e…
This paper presents a preliminary study comparing different observation and action space representations for Deep Reinforcement Learning (DRL) in the context of Real-time Strategy (RTS) games. Specifically, we compare two representations: (1) a global representation where the observation represents the whole game state…
Unsupervised learning representations generalize better than supervised learning under distribution shifts.