Multi-Loss Weighting with Coefficient of Variations

Groenendijk, Rick; Karaoğlu, Sezer; Gevers, Theo; Mensink, Thomas

doi:10.1109/wacv48630.2021.00151

Cited by 37 publications

(23 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For position and orientation specifically, notice how their computation is completely independent. This allows us optimize their losses L P and L R separately, bypassing the issue where their units are different and thus require re-weighting [5]. This also allows us to train both networks with different sets of data, which is necessary since the behavior of the position and orientation networks should not be correlated.…”

Section: A Training With Synthetic Datamentioning

confidence: 99%

Learning from Physical Human Feedback: An Object-Centric One-Shot Adaptation Method

Alvin¹,

Chen²,

Liu³

2022

Preprint

View full text Add to dashboard Cite

For robots to be effectively deployed in novel environments and tasks, they must be able to understand the feedback expressed by humans during intervention. This can either correct undesirable behavior or indicate additional preferences. Existing methods either require repeated episodes of interactions or assume prior known reward features, which is data-inefficient and can hardly transfer to new tasks. We relax these assumptions by describing human tasks in terms of objectcentric sub-tasks and interpreting physical interventions in relation to specific objects. Our method, Object Preference Adaptation (OPA), is composed of two key stages: 1) pre-training a base policy to produce a wide variety of behaviors, and 2) online-updating only certain weights in the model according to human feedback. The key to our fast, yet simple adaptation is that general interaction dynamics between agents and objects are fixed, and only object-specific preferences are updated. Our adaptation occurs online, requires only one human intervention (one-shot), and produces new behaviors never seen during training. Trained on cheap synthetic data instead of expensive human demonstrations, our policy demonstrates impressive adaptation to human perturbations on challenging, realistic tasks in our user study. Videos, code, and supplementary material: https://alvinosaur.github.io/AboutMe/projects/opa.

show abstract

Section: A Training With Synthetic Datamentioning

confidence: 99%

Learning from Physical Human Feedback: An Object-Centric One-Shot Adaptation Method

Alvin¹,

Chen²,

Liu³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In order to evaluate the benefits of combinations of two or more alignments, we employ the Multi-Loss Weighting with Coefficient of Variations (Groenendijk et al, 2021) technique (CoV) to calculate a weighted sum of auxiliary losses (Aux) that we add to the main XNLU losses L ic and L ec as follows:…”

Section: Adaptive Weighting Of Auxiliary Lossesmentioning

confidence: 99%

CrossAligner & Co: Zero-Shot Transfer Methods for Task-Oriented Cross-lingual Natural Language Understanding

Gritta¹,

Hu²,

Iacobacci³

2022

Preprint

View full text Add to dashboard Cite

Task-oriented personal assistants enable people to interact with a host of devices and services using natural language. One of the challenges of making neural dialogue systems available to more users is the lack of training data for all but a few languages. Zero-shot methods try to solve this issue by acquiring task knowledge in a high-resource language such as English with the aim of transferring it to the low-resource language(s). To this end, we introduce CrossAligner, the principal method of a variety of effective approaches for zero-shot cross-lingual transfer based on learning alignment from unlabelled parallel data. We present a quantitative analysis of individual methods as well as their weighted combinations, several of which exceed state-of-theart (SOTA) scores as evaluated across nine languages, fifteen test sets and three benchmark multilingual datasets. A detailed qualitative error analysis of the best methods shows that our fine-tuned language models can zero-shot transfer the task knowledge better than anticipated. * Work conducted as Research Intern at Huawei's Noah's Ark Lab, London. † -Equal contribution.1 Not to be confused with Lample and Conneau (2019).

show abstract

“…Uncertainty is generally used in three main contexts. Firstly, it is used as weight for loss or prediction reweighting [20,35], where the contribution of each part of the loss function or each prediction is weighted based on the certainty level of each task (multi-task learning) [34], or each loss function (single task learning) [20], or each prediction [50]. Secondly, uncertainty is treated as guidance for pseudo label quality estimation for weakly/semisupervised learning [55,80].…”

Section: Uncertainty Related Applicationsmentioning

confidence: 99%

Dense Uncertainty Estimation via an Ensemble-based Conditional Latent Variable Model

Zhang¹,

Dai²,

Harandi³

et al. 2021

Preprint

View full text Add to dashboard Cite

Uncertainty estimation has been extensively studied in recent literature, which can usually be classified as aleatoric uncertainty and epistemic uncertainty. In current aleatoric uncertainty estimation frameworks, it is often neglected that the aleatoric uncertainty is an inherent attribute of the data and can only be correctly estimated with an unbiased oracle model. Since the oracle model is inaccessible in most cases, we propose a new sampling and selection strategy at train time to approximate the oracle model for aleatoric uncertainty estimation. Further, we show a trivial solution in the dual-head based heteroscedastic aleatoric uncertainty estimation framework and introduce a new uncertainty consistency loss to avoid it. For epistemic uncertainty estimation, we argue that the internal variable in a conditional latent variable model is another source of epistemic uncertainty to model the predictive distribution and explore the limited knowledge about the hidden true model. We validate our observation on a dense prediction task, i.e., camouflaged object detection. Our results show that our solution achieves both accurate deterministic results and reliable uncertainty estimation.

show abstract

Multi-Loss Weighting with Coefficient of Variations

Cited by 37 publications

References 31 publications

Learning from Physical Human Feedback: An Object-Centric One-Shot Adaptation Method

Learning from Physical Human Feedback: An Object-Centric One-Shot Adaptation Method

CrossAligner & Co: Zero-Shot Transfer Methods for Task-Oriented Cross-lingual Natural Language Understanding

Dense Uncertainty Estimation via an Ensemble-based Conditional Latent Variable Model

Contact Info

Product

Resources

About