Publications

Monocular Robot Navigation with Self-Supervised Pretrained Vision Transformers

Miguel Saavedra-Ruiz

In this work, we consider the problem of learning a perception model for monocular robot navigation using few annotated images. Using a Visi… (see more)on Transformer (ViT) pretrained with a label-free self-supervised method, we successfully train a coarse image segmentation model for the Duckietown environment using 70 training images. Our model performs coarse image segmentation at the 8x8 patch level, and the inference resolution can be adjusted to balance prediction granularity and real-time perception constraints. We study how best to adapt a ViT to our task and environment, and find that some lightweight architectures can yield good single-image segmentation at a usable frame rate, even on CPU. The resulting perception model is used as the backbone for a simple yet robust visual servoing agent, which we deploy on a differential drive mobile robot to perform two tasks: lane following and obstacle avoidance.

2022-06-01

2022 19th Conference on Robots and Vision (CRV) (published)

doi.org

arxiv.org

Ageism and Artificial Intelligence: Protocol for a Scoping Review

Charlene H Chu

Kathleen Leslie

Jiamin Shi

Rune Nyrup

Andria Bianchi

Shehroz S Khan

S. A. Rahimi

Alexandra Lyn

Amanda Grenier

Background Artificial intelligence (AI) has emerged as a major driver of technological development in the 21st century, yet little attention… (see more) has been paid to algorithmic biases toward older adults. Objective This paper documents the search strategy and process for a scoping review exploring how age-related bias is encoded or amplified in AI systems as well as the corresponding legal and ethical implications. Methods The scoping review follows a 6-stage methodology framework developed by Arksey and O’Malley. The search strategy has been established in 6 databases. We will investigate the legal implications of ageism in AI by searching grey literature databases, targeted websites, and popular search engines and using an iterative search strategy. Studies meet the inclusion criteria if they are in English, peer-reviewed, available electronically in full text, and meet one of the following two additional criteria: (1) include “bias” related to AI in any application (eg, facial recognition) and (2) discuss bias related to the concept of old age or ageism. At least two reviewers will independently conduct the title, abstract, and full-text screening. Search results will be reported using the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) reporting guideline. We will chart data on a structured form and conduct a thematic analysis to highlight the societal, legal, and ethical implications reported in the literature. Results The database searches resulted in 7595 records when the searches were piloted in November 2021. The scoping review will be completed by December 2022. Conclusions The findings will provide interdisciplinary insights into the extent of age-related bias in AI systems. The results will contribute foundational knowledge that can encourage multisectoral cooperation to ensure that AI is developed and deployed in a manner consistent with ethical values and human rights legislation as it relates to an older and aging population. We will publish the review findings in peer-reviewed journals and disseminate the key results with stakeholders via workshops and webinars. Trial Registration OSF Registries AMG5P; https://osf.io/amg5p International Registered Report Identifier (IRRID) DERR1-10.2196/33211

2022-05-31

JMIR Research Protocols (published)

doi.org

Aligning artificial intelligence with climate change mitigation

Lynn H. Kaack

Priya L. Donti

Emma Strubell

George Kamiya

Felix Creutzig

David Rolnick

2022-05-31

Nature Climate Change (unknown)

doi.org

Conjugate Adder Net (CAddNet) - a Space-Efficient Approximate CNN

Lulan Shen

Maryam Ziaeefard

Brett Meyer

Warren Gross

James J. Clark

The AdderNet was recently developed as a way to implement deep neural networks without needing multiplication operations to combine weights … (see more)and inputs. Instead, absolute values of the difference between weights and inputs are used, greatly reducing the gate-level implementation complexity. Training of AdderNets is challenging, however, and the loss curves during training tend to fluctuate significantly. In this paper we propose the Conjugate Adder Network, or CAddNet, which uses the difference between the absolute values of conjugate pairs of inputs and the weights. We show that this can be implemented simply via a single minimum operation, resulting in a roughly 50% reduction in logic gate complexity as compared with AdderNets. The CAddNet method also stabilizes training as compared with AdderNets, yielding training curves similar to standard CNNs.

2022-05-31

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (published)

doi.org

Gradient-based learning drives robust representations in recurrent neural networks by balancing compression and expansion.

Matthew Farrell

Stefano Recanatesi

Timothy Moore

Guillaume Lajoie

Eric Shea-Brown

Neural networks need the right representations of input data to learn. Here we ask how gradient-based learning shapes a fundamental property… (see more) of representations in recurrent neural networks (RNNs)—their dimensionality. Through simulations and mathematical analysis, we show how gradient descent can lead RNNs to compress the dimensionality of their representations in a way that matches task demands during training while supporting generalization to unseen examples. This can require an expansion of dimensionality in early timesteps and compression in later ones, and strongly chaotic RNNs appear particularly adept at learning this balance. Beyond helping to elucidate the power of appropriately initialized artificial RNNs, this fact has implications for neurobiology as well. Neural circuits in the brain reveal both high variability associated with chaos and low-dimensional dynamical structures. Taken together, our findings show how simple gradient-based learning rules lead neural networks to solve tasks with robust representations that generalize to new cases. Neural networks in the brain often exhibit chaotic dynamics that can be captured by a small number of dimensions. Farrell et al. find that recurrent neural networks trained with gradient-based learning rules exhibit similar features. This helps form robust but generalizable input representations.

2022-05-31

Nature Machine Intelligence (published)

doi.org

High-Throughput and Energy-Efficient VLSI Architecture for Ordered Reliability Bits GRAND

Syed Mohsin Abbas

Thibaud Tonnellier

Furkan Ercan

Marwan Jalaleddine

Warren J. Gross

Ultrareliable low-latency communication (URLLC), a major 5G new-radio (NR) use case, is the key enabler for applications with strict reliabi… (see more)lity and latency requirements. These applications necessitate the use of short-length and high-rate channel codes. Guessing random additive noise decoding (GRAND) is a recently proposed maximum likelihood (ML) decoding technique for these short-length and high-rate codes. Rather than decoding the received vector, GRAND tries to infer the noise that corrupted the transmitted codeword during transmission through the communication channel. As a result, GRAND can decode any code, structured or unstructured. GRAND has hard-input as well as soft-input variants. Among these variants, ordered reliability bits GRAND (ORBGRAND) is a soft-input variant that outperforms hard-input GRAND and is suitable for parallel hardware implementation. This work reports the first hardware architecture for ORBGRAND, which achieves an average throughput of up to 42.5 Gb/s for a code length of 128 at a target frame error rate (FER) of 10−7. Furthermore, the proposed hardware can be used to decode any code as long as the length and rate constraints are met. In comparison to the GRAND with ABandonment (GRANDAB), a hard-input variant of GRAND, the proposed architecture enhances decoding performance by at least 2 dB. When compared to the state-of-the-art fast dynamic successive cancellation flip decoder (Fast-DSCF) using a 5G polar code (PC) (128, 105), the proposed ORBGRAND VLSI implementation has

2022-05-31

IEEE Transactions on Very Large Scale Integration (VLSI) Systems (published)

doi.org

arxiv.org

Optimal Control of Network-Coupled Subsystems: Spectral Decomposition and Low-Dimensional Solutions

Shuang Gao

Aditya Mahajan

In this article, we investigate the optimal control of network-coupled subsystems with coupled dynamics and costs. The dynamics coupling may… (see more) be represented by the adjacency matrix, the Laplacian matrix, or any other symmetric matrix corresponding to an underlying weighted undirected graph. Cost couplings are represented by two coupling matrices which have the same eigenvectors as the coupling matrix in the dynamics. We use the spectral decomposition of these three coupling matrices to decompose the overall system into

2022-05-31

IEEE Transactions on Control of Network Systems (published)

doi.org

arxiv.org

OptiMaP: swarm-powered Optimized 3D Mapping Pipeline for emergency response operations

Leandro R. Costa

Daniel Aloise

Luca G. Gianoli

Andrea Lodi

A smart application in sensing is mainly powered by a two-stage process comprising sensing (collect data) and computing (process data). Whil… (see more)e the sensing stage is typically performed locally through a dedicated Internet of Things infrastructure, the computing stage may require a powerful infrastructure in the cloud. However, when connectivity is poor and low latency becomes a requirement — as in emergency response and disaster relief operations — edge computing and ad hoc cloud paradigms come in support to keep the computing stage locally. Being local network connectivity and data processing limited, it is vital to properly optimize how the computing workload will be consumed by the local ad hoc cloud. For this purpose, we present and evaluate the swarm-powered Optimized 3D Mapping Pipeline (OptiMaP) for emergency response 3D mapping missions, which is implemented as a collaborative embedded Robot Operating System (ROS) application integrating an ad hoc telecommunication middleware.We simulate — with Software-In-The-Loop — realistic 3D mapping missions comprising up to 5 drones and 363 images covering 0.293km2. We show how the completion times of mapping missions carried out in a typical centralized manner can be dramatically reduced by two versions of the OptiMaP framework powered, respectively, by a variable neighborhood search heuristic and a greedy method.

2022-05-31

2022 18th International Conference on Distributed Computing in Sensor Systems (DCOSS) (published)

doi.org

Optimization and Simplification of PCPA Decoder for Reed-Muller Codes

Jiajie Li

Warren J. Gross

The collapsed projection-aggregation (CPA) decoder reduces the computational complexity of the recursive projection-aggregation (RPA) decode… (see more)r by removing the recursive structure. From simulations, the CPA decoder has similar error-correction performance as the RPA decoder, when decoding Reed-Muller (RM) (7, 3) and (8, 2) codes. The computational complexity can be further reduced by only selecting a subset of sub-spaces, which is achieved by pruning CPA decoders. In this work, optimization methods are proposed to find the pruned CPA (PCPA) decoder with small performance loss. Furthermore, the min-sum approximation is used to replace non-linear projection and aggregation functions, and a simplified list decoder based on the syndrome check is proposed. Under the same complexity, the optimized PCPA decoder has less performance loss than randomly constructed PCPA decoders in most case. The min-sum approximation incurs less than 0.15 dB performance loss at a target frame error rate of 10−4, and the simplified list decoder does not have noticeable performance loss.

2022-05-31

IEEE Communications Letters (published)

doi.org

Parametric Scattering Networks

Shanel Gauthier

Benjamin Therien

Laurent Alsène-Racicot

Michael Eickenberg

The wavelet scattering transform creates geometric invariants and deformation stability. In multiple signal domains, it has been shown to yi… (see more)eld more discriminative representations compared to other non-learned representations and to outperform learned representations in certain tasks, particularly on limited labeled data and highly structured signals. The wavelet filters used in the scattering transform are typically selected to create a tight frame via a parameterized mother wavelet. In this work, we investigate whether this standard wavelet filterbank construction is optimal. Focusing on Morlet wavelets, we propose to learn the scales, orientations, and aspect ratios of the filters to produce problem-specific parameterizations of the scattering transform. We show that our learned versions of the scattering transform yield significant performance gains in small-sample classification settings over the standard scattering transform. Moreover, our empirical results suggest that traditional filterbank constructions may not always be necessary for scattering transforms to extract effective representations.

2022-05-31

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (published)

doi.org

arxiv.org

Towards an AAK Theory Approach to Approximate Minimization in the Multi-Letter Case

Clara Lacroce

Prakash Panangaden

Guillaume Rabusseau

We study the approximate minimization problem of weighted finite automata (WFAs): given a WFA, we want to compute its optimal approximation … (see more)when restricted to a given size. We reformulate the problem as a rank-minimization task in the spectral norm, and propose a framework to apply Adamyan-Arov-Krein (AAK) theory to the approximation problem. This approach has already been successfully applied to the case of WFAs and language modelling black boxes over one-letter alphabets \citep{AAK-WFA,AAK-RNN}. Extending the result to multi-letter alphabets requires solving the following two steps. First, we need to reformulate the approximation problem in terms of noncommutative Hankel operators and noncommutative functions, in order to apply results from multivariable operator theory. Secondly, to obtain the optimal approximation we need a version of noncommutative AAK theory that is constructive. In this paper, we successfully tackle the first step, while the second challenge remains open.

2022-05-31

ArXiv (preprint)

doi.org

arxiv.org

Evolving Domain Generalization

Wei Wang

Gezheng Xu

Ruizhi Pu

Jiaqi Li

Fan Zhou

Changjian Shui

Charles Ling