Publications

ArK: Augmented Reality with Knowledge Emergent Infrastructure
Qiuyuan Huang
J. Park
Abhinav Gupta
Pan Lu
Paul N. Bennett
Ran Gong
Subhojit Som
Baolin Peng
Owais Khan Mohammed
Christopher Pal
Yejin Choi
Jianfeng Gao
Despite the growing adoption of mixed reality and interactive AI, it remains challenging to generate high-quality 2D/3D scenes in unseen env… (voir plus)ironments. Typically, an AI agent requires collecting extensive training data for every new task, which can be costly or impossible for many domains. In this study, we develop an infinite agent that learns to transfer knowledge memory from general foundation models (e.g., GPT4, DALLE) to novel domains or scenarios for scene understanding and generation in physical or virtual worlds. Central to our approach is the interactive emerging mechanism, dubbed Augmented Reality with Knowledge Emergent Infrastructure (ArK) , which leverages knowledge-memory to generate scenes in unseen physical worlds and virtual reality environments. The knowledge interactive emergent ability (Figure 1) is demonstrated through i) micro-action of cross-modality : in multi-modality models to collect a large amount of relevant knowledge-memory data for each interaction task (e.g., unseen scene understanding) from the physical reality; and ii) macro-behavior of reality-agnostic : in mix-reality environments to improve interactions that tailor to different characterized roles, target variables, collaborative information, and so on. We validate ArK’s effectiveness in scene generation and editing tasks and show that our ArK approach, combined with large foundation models, significantly improves the quality of generated 2D/3D scenes, highlighting its potential in applications such as metaverse and gaming simulation.
2022-12-31
(publié)
www.semanticscholar.org
Augmenting Transit Network Design Algorithms with Deep Learning
Andrew Holliday
Gregory Dudek
This paper considers the use of deep learning models to enhance optimization algorithms for transit network design. Transit network design i… (voir plus)s the problem of determining routes for transit vehicles that minimize travel time and operating costs, while achieving full service coverage. State-of-the-art meta-heuristic search algorithms give good results on this problem, but can be very time-consuming. In contrast, neural networks can learn sub-optimal but fast-to-compute heuristics based on large amounts of data. Combining these approaches, we develop a fast graph neural network model for transit planning, and use it to initialize state-of-the-art search algorithms. We show that this combination can improve the results of these algorithms on a variety of metrics by up to 17%, without increasing their run time; or they can match the quality of the original algorithms while reducing the computing time by up to a factor of 50.
2022-12-31
2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC) (publié)
doi.org
Auxiliary Losses for Learning Generalizable Concept-based Models
Ivaxi Sheth
S Ebrahimi Kahou
2022-12-31
Advances in Neural Information Processing Systems 36 (NeurIPS 2023) (publié)
doi.org
openreview.net
Bayes-MIL: A New Probabilistic Perspective on Attention-based Multiple Instance Learning for Whole Slide Images
Yufei Cui
Ziquan Liu
Xiangyu Liu
Xue Liu
Cong Wang
Tei-Wei Kuo
Chun Jason Xue
Antoni B. Chan
Multiple instance learning (MIL) is a popular weakly-supervised learning model on the whole slide image (WSI) for AI-assisted pathology diag… (voir plus)nosis. The recent advance in attention-based MIL allows the model to find its region-of-interest (ROI) for interpretation by learning the attention weights for image patches of WSI slides. However, we empirically find that the interpretability of some related methods is either untrustworthy as the principle of MIL is violated or unsatisfactory as the high-attention regions are not consistent with experts’ annotations. In this paper, we propose Bayes-MIL to address the problem from a probabilistic perspective. The induced patch-level uncertainty is proposed as a new measure of MIL interpretability, which outperforms previous methods in matching doctors annotations. We design a slide-dependent patch regularizer (SDPR) for the attention, imposing constraints derived from the MIL assumption, on the attention distribution. SDPR explicitly constrains the model to generate correct attention values. The spatial information is further encoded by an approximate convolutional conditional random field (CRF), for better interpretability. Experimental results show Bayes-MIL outperforms the related methods in patch-level and slide-level metrics and provides much better interpretable ROI on several large-scale WSI datasets.
2022-12-31
ICLR (publié)
dblp.uni-trier.de
Benchmarking Graph Neural Networks
Vijay Prakash Dwivedi
Chaitanya K. Joshi
Thomas Laurent
Anh Tuan Luu
Yoshua Bengio
Xavier Bresson
2022-12-31
J. Mach. Learn. Res. (publié)
arxiv.org
arxiv.org
Benchmarking State-Merging Algorithms for Learning Regular Languages.
Adil Soubki
Jeffrey Heinz
François Coste
Faissal Ouardi
Guillaume Rabusseau
2022-12-31
International Conference on Graphics and Interaction (publié)
dblp.uni-trier.de
Best-Case Retrieval Evaluation: Improving the Sensitivity of Reciprocal Rank with Lexicographic Precision
Fernando Diaz
Across a variety of ranking tasks, researchers use reciprocal rank to measure the effectiveness for users interested in exactly one relevant… (voir plus) item. Despite its widespread use, evidence suggests that reciprocal rank is brittle when discriminating between systems. This brittleness, in turn, is compounded in modern evaluation settings where current, high-precision systems may be difficult to distinguish. We address the lack of sensitivity of reciprocal rank by introducing and connecting it to the concept of best-case retrieval, an evaluation method focusing on assessing the quality of a ranking for the most satisfied possible user across possible recall requirements. This perspective allows us to generalize reciprocal rank and define a new preference-based evaluation we call lexicographic precision or lexiprecision. By mathematical construction, we ensure that lexiprecision preserves differences detected by reciprocal rank, while empirically improving sensitivity and robustness across a broad set of retrieval and recommendation tasks.
2022-12-31
EVIA (publié)
doi.org
arxiv.org
Block-State Transformers
Mahan Fathi
Jonathan Pilault
Pierre-Luc Bacon
Christopher Pal
Orhan Firat
Ross Goroshin
State space models (SSMs) have shown impressive results on tasks that require modeling long-range dependencies and efficiently scale to long… (voir plus) sequences owing to their subquadratic runtime complexity. Originally designed for continuous signals, SSMs have shown superior performance on a plethora of tasks, in vision and audio; however, SSMs still lag Transformer performance in Language Modeling tasks. In this work, we propose a hybrid layer named Block-State Transformer (BST), that internally combines an SSM sublayer for long-range contextualization, and a Block Transformer sublayer for short-term representation of sequences. We study three different, and completely parallelizable, variants that integrate SSMs and block-wise attention. We show that our model outperforms similar Transformer-based architectures on language modeling perplexity and generalizes to longer sequences. In addition, the Block-State Transformer demonstrates more than tenfold increase in speed at the layer level compared to the Block-Recurrent Transformer when model parallelization is employed.
2022-12-31
arXiv (prépublication)
doi.org
openreview.net
Bugs in the Data: How ImageNet Misrepresents Biodiversity
Alexandra Luccioni
David Rolnick
ImageNet-1k is a dataset often used for benchmarking machine learning (ML) models and evaluating tasks such as image recognition and object … (voir plus)detection. Wild animals make up 27% of ImageNet-1k but, unlike classes representing people and objects, these data have not been closely scrutinized. In the current paper, we analyze the 13,450 images from 269 classes that represent wild animals in the ImageNet-1k validation set, with the participation of expert ecologists. We find that many of the classes are ill-defined or overlapping, and that 12% of the images are incorrectly labeled, with some classes having >90% of images incorrect. We also find that both the wildlife-related labels and images included in ImageNet-1k present significant geographical and cultural biases, as well as ambiguities such as artificial animals, multiple species in the same image, or the presence of humans. Our findings highlight serious issues with the extensive use of this dataset for evaluating ML systems, the use of such algorithms in wildlife-related tasks, and more broadly the ways in which ML datasets are commonly created and curated.
2022-12-31
AAAI (publié)
doi.org
arxiv.org
Cache-Efficient Dynamic Programming MDP Solver
Jaël Champagne Gareau
Guillaume Gosset
Éric Beaudry
Vladimir Makarenkov
2022-12-31
European Conference on Artificial Intelligence (publié)
doi.org
Can Ensembling Pre-processing Algorithms Lead to Better Machine Learning Fairness?
Khaled Badran
Pierre-Olivier Côté
Amanda Kolopanis
Rached Bouchoucha
Antonio Collante
Diego Elias Costa
Emad Shihab
Foutse Khomh
As machine learning (ML) systems get adopted in more critical areas, it has become increasingly crucial to address the bias that could occur… (voir plus) in these systems. Several fairness pre-processing algorithms are available to alleviate implicit biases during model training. These algorithms employ different concepts of fairness, often leading to conflicting strategies with consequential trade-offs between fairness and accuracy. In this work, we evaluate three popular fairness pre-processing algorithms and investigate the potential for combining all algorithms into a more robust pre-processing ensemble. We report on lessons learned that can help practitioners better select fairness algorithms for their models.
2022-12-31
Computer (publié)
doi.org
arxiv.org
Can Forward Gradient Match Backpropagation?
Louis Fournier
Stéphane Rivaud
Eugene Belilovsky
Michael Eickenberg
Edouard Oyallon
Forward Gradients - the idea of using directional derivatives in forward differentiation mode - have recently been shown to be utilizable fo… (voir plus)r neural network training while avoiding problems generally associated with backpropagation gradient computation, such as locking and memorization requirements. The cost is the requirement to guess the step direction, which is hard in high dimensions. While current solutions rely on weighted averages over isotropic guess vector distributions, we propose to strongly bias our gradient guesses in directions that are much more promising, such as feedback obtained from small, local auxiliary networks. For a standard computer vision neural network, we conduct a rigorous study systematically covering a variety of combinations of gradient targets and gradient guesses, including those previously presented in the literature. We find that using gradients obtained from a local loss as a candidate direction drastically improves on random noise in Forward Gradient methods.
2022-12-31
ICML (publié)
doi.org
proceedings.mlr.press