Active Ordinal Querying for Tuplewise Similarity Learning

Canal, Gregory; Fenu, Stefano; Rozell, Christopher J.

doi:10.1609/aaai.v34i04.5734

Cited by 4 publications

(21 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ranking. In the second simulation, we compare the performance of actively selected nearest neighbor queries against ranking queries [10]. We observe that nearest neighbor queries perform competitively to ranking queries, as illustrated in Fig.…”

Section: C2 Mds Embedding Learningmentioning

confidence: 97%

“…where D Qn refers to the set of distances between the reference r n and each of t c n ∈ T n and N n−1 Qn represents the assumed normal distribution on D Qn with the mean and variance determined by the estimate of distances after n − 1 queries. Due to this normal distribution assumption, the entropy in (10) and the expectation in (11) are straightforward calculations. The full procedure is shown in Alg.…”

Section: B Computation Of Mutual Informationmentioning

confidence: 99%

“…We perform a set of experiments which utilize MDS to learn representations of the items. In particular, we use this opportunity to compare the performance of NN queries against a more complex ranking query [10]. When comparing against ranking queries, it is important to note that we expect both actively selected and randomly selected ranking queries to outperform a nearest neighbor query of the same size on a per-query basis, as there is a discrepancy in the amount of information each query contains.…”

Section: C2 Mds Embedding Learningmentioning

confidence: 99%

“…For the following experiments, Info-Ranking-C means the active selection method in [10] was used to select ranking queries each with a set T i of size C.…”

Section: C2 Mds Embedding Learningmentioning

confidence: 99%

“…To the best of our knowledge, we are the first to study this query type. Similar ideas have been explored before, such as using UI configurations to collect multiple triplets at once [8], enforcing a class-similarity based quadruplet loss (one anchor, one positive point, two negative points) [9], and soliciting ranking queries [10]. Of these approaches, [9] is the most similar, but NN queries are 1) not confined to a particular fixed length, and 2) not restricted to using class information.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Active metric learning and classification using similarity queries

Nadagouda¹,

Xu²,

Davenport³

2022

Preprint

View full text Add to dashboard Cite

Active learning is commonly used to train label-efficient models by adaptively selecting the most informative queries. However, most active learning strategies are designed to either learn a representation of the data (e.g., embedding or metric learning) or perform well on a task (e.g., classification) on the data. However, many machine learning tasks involve a combination of both representation learning and a task-specific goal. Motivated by this, we propose a novel unified query framework that can be applied to any problem in which a key component is learning a representation of the data that reflects similarity. Our approach builds on similarity or nearest neighbor (NN) queries which seek to select samples that result in improved embeddings. The queries consist of a reference and a set of objects, with an oracle selecting the object most similar (i.e., nearest) to the reference. In order to reduce the number of solicited queries, they are chosen adaptively according to an information theoretic criterion. We demonstrate the effectiveness of the proposed strategy on two tasks -active metric learning and active classificationusing a variety of synthetic and real world datasets. In particular, we demonstrate that actively selected NN queries outperform recently developed active triplet selection methods in a deep metric learning setting. Further, we show that in classification, actively selecting class labels can be reformulated as a process of selecting the most informative NN query, allowing direct application of our method.

show abstract

Section: C2 Mds Embedding Learningmentioning

confidence: 97%

Section: B Computation Of Mutual Informationmentioning

confidence: 99%

Section: C2 Mds Embedding Learningmentioning

confidence: 99%

“…For the following experiments, Info-Ranking-C means the active selection method in [10] was used to select ranking queries each with a set T i of size C.…”

Section: C2 Mds Embedding Learningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Active metric learning and classification using similarity queries

Nadagouda¹,

Xu²,

Davenport³

2022

Preprint

View full text Add to dashboard Cite

show abstract

Is subjective perceptual similarity metacognitive?

Moharramipour,

Zhou,

Rahnev

et al. 2024

Preprint

View full text Add to dashboard Cite

Perceptual similarity is a cornerstone for human learning and generalization. However, in assessing the similarity between two stimuli differing in multiple dimensions, it is not well-defined which feature(s) one should focus on. The problem has accordingly been considered ill-posed. We hypothesize that similarity judgments may be, in a sense, metacognitive: The stimuli rated as subjectively similar are those that are in fact more challenging for oneself to discern in practice, in near-threshold settings (e.g., psychophysics experiments). This self-knowledge about one’s own perceptual capacities provides a quasi-objective ground truth as to whether two stimuli ‘should’ be judged as similar. To test this idea, we measure perceptual discrimination capacity between face pairs, and ask subjects to rank the similarity between them. Based on pilot data, we hypothesize a positive association between perceptual discrimination capacity and subjective dissimilarity, with this association being importantly specific to each individual.

show abstract

One for All: Simultaneous Metric and Preference Learning over Multiple Users

Canal¹,

Mason²,

Vinayak³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

This paper investigates simultaneous preference and metric learning from a crowd of respondents. A set of items represented by d-dimensional feature vectors and paired comparisons of the form "item i is preferable to item j" made by each user is given. Our model jointly learns a distance metric that characterizes the crowd's general measure of item similarities along with a latent ideal point for each user reflecting their individual preferences. This model has the flexibility to capture individual preferences, while enjoying a metric learning sample cost that is amortized over the crowd. We first study this problem in a noiseless, continuous response setting (i.e., responses equal to differences of item distances) to understand the fundamental limits of learning. Next, we establish prediction error guarantees for noisy, binary measurements such as may be collected from human respondents, and show how the sample complexity improves when the underlying metric is lowrank. Finally, we establish recovery guarantees under assumptions on the response distribution. We demonstrate the performance of our model on both simulated data and on a dataset of color preference judgements across a large number of users.Preprint. Under review.

show abstract

Active Ordinal Querying for Tuplewise Similarity Learning

Cited by 4 publications

References 24 publications

Active metric learning and classification using similarity queries

Active metric learning and classification using similarity queries

Is subjective perceptual similarity metacognitive?

One for All: Simultaneous Metric and Preference Learning over Multiple Users

Contact Info

Product

Resources

About