Nous utilisons des témoins pour analyser le trafic et l’utilisation de notre site web, afin de personnaliser votre expérience. Vous pouvez désactiver ces technologies à tout moment, mais cela peut restreindre certaines fonctionnalités du site. Consultez notre Politique de protection de la vie privée pour en savoir plus.
Paramètre des cookies
Vous pouvez activer et désactiver les types de cookies que vous souhaitez accepter. Cependant certains choix que vous ferez pourraient affecter les services proposés sur nos sites (ex : suggestions, annonces personnalisées, etc.).
Cookies essentiels
Ces cookies sont nécessaires au fonctionnement du site et ne peuvent être désactivés. (Toujours actif)
Cookies analyse
Acceptez-vous l'utilisation de cookies pour mesurer l'audience de nos sites ?
Multimedia Player
Acceptez-vous l'utilisation de cookies pour afficher et vous permettre de regarder les contenus vidéo hébergés par nos partenaires (YouTube, etc.) ?
Publications
Deep Discourse Analysis for Generating Personalized Feedback in Intelligent Tutor Systems
We explore creating automated, personalized feedback in an intelligent tutoring system (ITS). Our goal is to pinpoint correct and incorrect … (voir plus)concepts in student answers in order to achieve better student learning gains. Although automatic methods for providing personalized feedback exist, they do not explicitly inform students about which concepts in their answers are correct or incorrect. Our approach involves decomposing students answers using neural discourse segmentation and classification techniques. This decomposition yields a relational graph over all discourse units covered by the reference solutions and student answers. We use this inferred relational graph structure and a neural classifier to match student answers with reference solutions and generate personalized feedback. Although the process is completely automated and data-driven, the personalized feedback generated is highly contextual, domain-aware and effectively targets each student's misconceptions and knowledge gaps. We test our method in a dialogue-based ITS and demonstrate that our approach results in high-quality feedback and significantly improved student learning gains.
2021-05-18
Proceedings of the AAAI Conference on Artificial Intelligence (publié)
AI Safety is a major concern in many deep learning applications such as autonomous driving. Given a trained deep learning model, an importan… (voir plus)t natural problem is how to reliably verify the model's prediction. In this paper, we propose a novel framework --- deep verifier networks (DVN) to detect unreliable inputs or predictions of deep discriminative models, using separately trained deep generative models. Our proposed model is based on conditional variational auto-encoders with disentanglement constraints to separate the label information from the latent representation. We give both intuitive and theoretical justifications for the model. Our verifier network is trained independently with the prediction model, which eliminates the need of retraining the verifier network for a new model. We test the verifier network on both out-of-distribution detection and adversarial example detection problems, as well as anomaly detection problems in structured prediction tasks such as image caption generation. We achieve state-of-the-art results in all of these problems.
2021-05-18
Proceedings of the AAAI Conference on Artificial Intelligence (publié)
Can meta-learning discover generic ways of processing time series (TS) from a diverse dataset so as to greatly improve generalization on new… (voir plus) TS coming from different datasets? This work provides positive evidence to this using a broad meta-learning framework which we show subsumes many existing meta-learning algorithms. Our theoretical analysis suggests that residual connections act as a meta-learning adaptation mechanism, generating a subset of task-specific parameters based on a given TS input, thus gradually expanding the expressive power of the architecture on-the-fly. The same mechanism is shown via linearization analysis to have the interpretation of a sequential update of the final linear layer. Our empirical results on a wide range of data emphasize the importance of the identified meta-learning mechanisms for successful zero-shot univariate forecasting, suggesting that it is viable to train a neural network on a source TS dataset and deploy it on a different target TS dataset without retraining, resulting in performance that is at least as good as that of state-of-practice univariate forecasting models.
2021-05-18
Proceedings of the AAAI Conference on Artificial Intelligence (publié)
Branch and Bound (B&B) is the exact tree search method typically used to solve Mixed-Integer Linear Programming problems (MILPs). Learning b… (voir plus)ranching policies for MILP has become an active research area, with most works proposing to imitate the strong branching rule and specialize it to distinct classes of problems. We aim instead at learning a policy that generalizes across heterogeneous MILPs: our main hypothesis is that parameterizing the state of the B&B search tree can aid this type of generalization. We propose a novel imitation learning framework, and introduce new input features and architectures to represent branching. Experiments on MILP benchmark instances clearly show the advantages of incorporating an explicit parameterization of the state of the search tree to modulate the branching decisions, in terms of both higher accuracy and smaller B&B trees. The resulting policies significantly outperform the current state-of-the-art method for "learning to branch" by effectively allowing generalization to generic unseen instances.
2021-05-18
Proceedings of the AAAI Conference on Artificial Intelligence (publié)
Visual saliency has emerged as a major visualization tool for interpreting deep reinforcement learning (RL) agents. However, much of the exi… (voir plus)sting research uses it as an analyzing tool rather than an inductive bias for policy learning. In this work, we use visual attention as an inductive bias for RL agents. We propose a novel self-supervised attention learning approach which can 1. learn to select regions of interest without explicit annotations, and 2. act as a plug for existing deep RL methods to improve the learning performance. We empirically show that the self-supervised attention-aware deep RL methods outperform the baselines in the context of both the rate of convergence and performance. Furthermore, the proposed self-supervised attention is not tied with specific policies, nor restricted to a specific scene. We posit that the proposed approach is a general self-supervised attention module for multi-task learning and transfer learning, and empirically validate the generalization ability of the proposed method. Finally, we show that our method learns meaningful object keypoints highlighting improvements both qualitatively and quantitatively.
2021-05-18
AAAI Conference on Artificial Intelligence (publié)
The search for appropriate treatments of cerebral palsy (CP) would be facilitated if researchers could non-invasively monitor anatomical cha… (voir plus)nges in the spinal cord. The study by Trevarrow et al. aims to validate the relevance of magnetization transfer ratio and diffusion tensor imaging, both magnetic resonance imaging (MRI) techniques, to quantify microstructural abnormalities in the spinal cord of adult patients with CP. The authors used a semi-automated atlas-based analysis pipeline based on Spinal Cord Toolbox software to compute cord and gray matter atrophy and to quantify MRI metrics in specific spinal tracts. In line with their hypothesis, Trevarrow et al. observed differences in cord and gray matter size between participants with CP and typically developing peers. Interestingly, they also demonstrated an association between these morphometric biomarkers and clinical scores of hand dexterity. Magnetization transfer ratio was also reduced in the corticospinal tract of patients with CP. The study by Trevarrow et al. is a remarkable tour de force in that it is extremely difficult to image patients with CP as they are prone to motion (spasticity). In particular, gradient-echo sequences, used for magnetization transfer imaging, are particularly sensitive to motion, as can be seen on Figure 1b of the article. Echo planar imaging sequences, used for diffusion imaging, are sensitive to magnetic field inhomogeneities, which are prevalent in the spine region. The authors used an MRI acquisition protocol similar to a recently proposed standardized quantitative spinal cord MRI protocol (https://spine-generic.rtfd.io/), which likely helped them to obtain satisfactory images despite the many aforementioned challenges. From an image analysis standpoint, one limitation associated with atlas-based analysis (acknowledged by the authors) is that the registration to the template only relies on the spinal cord contour, not its internal structure. In other words, the white matter atlas is adjusted to the participant’s spinal cord contour, and the internal structure of the cord is quasi-linearly scaled (based on a B-spline regularized deformation). This quasi-linearity assumption might not hold true if, for example, the gray/white matter ratio differs between the participant and the template, and/ or the spatial location of the white matter tracts differs between the participant and the atlas, and/or specific tracts (e.g. corticospinal) degenerate. All these effects would cause a mismatch between the warped atlas’ and the participant’s white matter tracts. Unfortunately, there is no solution to this problem (yet). There are ways, however, to mitigate it. For example, using imaging sequences that are sensitive to some internal structures of the spinal cord, such as gray matter, or even some white matter tracts. These internal structures could then be accounted for during registration. However, these advanced contrast techniques are themselves noisy and sensitive to participant motion. In conclusion, the study by Trevarrow et al. is a remarkable technical achievement and a concrete first step towards the inclusion of microstructure MRI to the assessment of spinal cord integrity in patients with CP. Limitations, mostly related to data acquisition, could be tackled with the development of better solutions for gradient echo sequences in participants that are prone to motion. Navigator and/or advanced shimming approaches will hopefully mitigate these issues, making spinal cord quantitative MRI more amenable to clinical routine.
