Artificial Intelligence
Making machines smarter for improved safety, efficiency and comfort.
Our AI research encompasses advances in computer vision, speech and audio processing, as well as data analytics. Key research themes include improved perception based on machine learning techniques, learning control policies through model-based reinforcement learning, as well as cognition and reasoning based on learned semantic representations. We apply our work to a broad range of automotive and robotics applications, as well as building and home systems.
Quick Links
-
Researchers
Jonathan
Le Roux
Toshiaki
Koike-Akino
Ye
Wang
Anoop
Cherian
Tim K.
Marks
Chiori
Hori
Michael J.
Jones
Gordon
Wichern
Daniel N.
Nikovski
Devesh K.
Jha
Kieran
Parsons
Philip V.
Orlik
Matthew
Brand
Suhas
Lohit
Diego
Romeres
Hassan
Mansour
William S.
Yerazunis
Petros T.
Boufounos
Mouhacine
Benosman
Arvind
Raghunathan
Siddarth
Jain
Kuan-Chuan
Peng
Hongbo
Sun
Yebin
Wang
Radu
Corcodel
Chungwei
Lin
Pu
(Perry)
Wang
Stefano
Di Cairano
Jianlin
Guo
Anthony
Vetro
Bingnan
Wang
Jinyun
Zhang
Jose
Amaya
Karl
Berntorp
Ankush
Chakrabarty
Marcus
Greiff
Dehong
Liu
Yanting
Ma
Kei
Ota
Wataru
Tsujita
Jing
Liu
Abraham P.
Vinod
-
Awards
-
AWARD ACM/IEEE Design Automation Conference 2022 Best Paper Award nominee Date: July 14, 2022
Awarded to: Weidong Cao, Mouhacine Benosman, Xuan Zhang, and Rui Ma
MERL Contact: Mouhacine Benosman
Research Area: Artificial IntelligenceBrief- The Conference committee of the 59th Design Automation Conference has chosen MERL's paper entitled 'Domain Knowledge-Infused Deep Learning for Automated Analog/RF Circuit Parameter Optimization', as a DAC Best Paper Award nominee. The committee evaluated both manuscript and submitted presentation recording, and has chosen MERL's paper as one of six nominees for this prestigious award. Decisions were based on the submissions’ innovation, impact and exposition.
-
AWARD International Conference on Artificial Intelligence Circuits and Systems (AICAS) 2022 Openedges Award Date: June 15, 2022
Awarded to: Yuxiang Sun, Mouhacine Benosman, Rui Ma.
MERL Contact: Mouhacine Benosman
Research Area: Artificial IntelligenceBrief- The committee of the International Conference on Artificial Intelligence Circuits and Systems (AICAS) 2022, has selected MERL's paper entitled 'GaN Distributed RF Power Amplifier Automation Design with Deep Reinforcement Learning' as a winner of the AICAS 2022 Openedges Award.
In this paper MERL researchers propose a novel design automation methodology based on deep reinforcement learning (RL), for wide-band non-uniform distributed RF power amplifiers, known for their high dimensional design challenges.
- The committee of the International Conference on Artificial Intelligence Circuits and Systems (AICAS) 2022, has selected MERL's paper entitled 'GaN Distributed RF Power Amplifier Automation Design with Deep Reinforcement Learning' as a winner of the AICAS 2022 Openedges Award.
-
AWARD MERL Ranked 1st Place in Cross-Subject Transfer Learning Task and 4th Place Overall at the NeurIPS2021 BEETL Competition for EEG Transfer Learning. Date: November 11, 2021
Awarded to: Niklas Smedemark-Margulies, Toshiaki Koike-Akino, Ye Wang, Deniz Erdogmus
MERL Contacts: Toshiaki Koike-Akino; Ye Wang
Research Areas: Artificial Intelligence, Signal Processing, Human-Computer InteractionBrief- The MERL Signal Processing group achieved first place in the cross-subject transfer learning task and fourth place overall in the NeurIPS 2021 BEETL AI Challenge for EEG Transfer Learning. The team included Niklas Smedemark-Margulies (intern from Northeastern University), Toshiaki Koike-Akino, Ye Wang, and Prof. Deniz Erdogmus (Northeastern University). The challenge addresses two types of transfer learning tasks for EEG Biosignals: a homogeneous transfer learning task for cross-subject domain adaptation; and a heterogeneous transfer learning task for cross-data domain adaptation. There were 110+ registered teams in this competition, MERL ranked 1st in the homogeneous transfer learning task, 7th place in the heterogeneous transfer learning task, and 4th place for the combined overall score. For the homogeneous transfer learning task, MERL developed a new pre-shot learning framework based on feature disentanglement techniques for robustness against inter-subject variation to enable calibration-free brain-computer interfaces (BCI). MERL is invited to present our pre-shot learning technique at the NeurIPS 2021 workshop.
See All Awards for Artificial Intelligence -
-
News & Events
-
TALK [MERL Seminar Series 2023] Dr. Suraj Srinivas presents talk titled Pitfalls and Opportunities in Interpretable Machine Learning Date & Time: Tuesday, March 14, 2023; 1:00 PM
Speaker: Suraj Srinivas, Harvard University
MERL Host: Suhas Lohit
Research Areas: Artificial Intelligence, Computer Vision, Machine LearningAbstract
In this talk, I will discuss our recent research on understanding post-hoc interpretability. I will begin by introducing a characterization of post-hoc interpretability methods as local function approximators, and the implications of this viewpoint, including a no-free-lunch theorem for explanations. Next, we shall challenge the assumption that post-hoc explanations provide information about a model's discriminative capabilities p(y|x) and instead demonstrate that many common methods instead rely on a conditional generative model p(x|y). This observation underscores the importance of being cautious when using such methods in practice. Finally, I will propose to resolve this via regularization of model structure, specifically by training low curvature neural networks, resulting in improved model robustness and stable gradients.
-
NEWS Jonathan Le Roux gives invited talk at CMU's Language Technology Institute Colloquium Date: December 9, 2022
Where: Pittsburg, PA
MERL Contact: Jonathan Le Roux
Research Areas: Artificial Intelligence, Machine Learning, Speech & AudioBrief- MERL Senior Principal Research Scientist and Speech and Audio Senior Team Leader, Jonathan Le Roux, was invited by Carnegie Mellon University's Language Technology Institute (LTI) to give an invited talk as part of the LTI Colloquium Series. The LTI Colloquium is a prestigious series of talks given by experts from across the country related to different areas of language technologies. Jonathan's talk, entitled "Towards general and flexible audio source separation", presented an overview of techniques developed at MERL towards the goal of robustly and flexibly decomposing and analyzing an acoustic scene, describing in particular the Speech and Audio Team's efforts to extend MERL's early speech separation and enhancement methods to more challenging environments, and to more general and less supervised scenarios.
See All News & Events for Artificial Intelligence -
-
Research Highlights
-
Internships
-
CV1938: Component transfer learning for RL and robotic applications
MERL is offering a new research internship opportunity in the field of Transfer Learning for Deep RL. The position requires a strong background in Deep RL, excellent programming skills and experience with robotics is preferred. The position is open to graduate students on a PhD track only, and the length of the internship is three months with the possibility of extending if required. The intern is expected to disseminate this research in top tier scientific conferences such as RSS, IROS, ICRA etc., and if applicable, help with filing associated patents. Start and end dates are flexible.
-
CI1946: Robust, Private, and Efficient Machine Learning
MERL is seeking highly motivated and qualified interns to work on fundamental machine learning techniques for robustness, privacy, and efficiency. The ideal candidates would have significant research experience in one or more of the following topics: robust machine learning methods, defenses against adversarial examples, privacy issues in machine learning, membership inference attacks, federated/distributed learning, and/or efficient/Green AI. A mature understanding of modern machine learning methods, proficiency with Python, and familiarity with deep learning frameworks are expected. Proficiency with other programming languages and software development experience is a plus. Candidates at or beyond the middle of their Ph.D. program are encouraged to apply. Multiple positions are available throughout 2023 (Spring/Summer of course, but also as early as January), with expected durations of 3-6 months and flexible start dates.
-
ST1750: THz (Terahertz) Sensing
The Signal Processing (SP) group at MERL is seeking a highly motivated intern to conduct fundamental research in THz (Terahertz) sensing. Expertise in statistical inference, unsupervised anomaly detection, and deep learning (spatial-temporal representation learning) is required. Previous hands-on experience in THz data analysis is a plus. Familiarity with python and deep learning libraries is a must. The intern will collaborate with a small group of MERL researchers to develop novel algorithms, design experiments with collaborators, and prepare results for patents and publication. The expected duration of the internship is 3 months with a flexible start date.
See All Internships for Artificial Intelligence -
-
Recent Publications
- , "Are Deep Neural Networks SMARTer than Second Graders?", IEEE Conference on Computer Vision and Pattern Recognition (CVPR), March 2023.BibTeX TR2023-014 PDF Data
- @inproceedings{Cherian2023mar2,
- author = {Cherian, Anoop and Peng, Kuan-Chuan and Lohit, Suhas and Smith, Kevin and Tenenbaum, Joshua B.},
- title = {Are Deep Neural Networks SMARTer than Second Graders?},
- booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
- year = 2023,
- month = mar,
- url = {https://www.merl.com/publications/TR2023-014}
- }
- , "Discriminative 3D Shape Modeling for Few-Shot Instance Segmentation", IEEE International Conference on Robotics and Automation (ICRA), March 2023.BibTeX TR2023-010 PDF
- @inproceedings{Cherian2023mar,
- author = {Cherian, Anoop and Jain, Siddarth and Marks, Tim K. and Sullivan, Alan},
- title = {Discriminative 3D Shape Modeling for Few-Shot Instance Segmentation},
- booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
- year = 2023,
- month = mar,
- url = {https://www.merl.com/publications/TR2023-010}
- }
- , "H-SAUR: Hypothesize, Simulate, Act, Update, and Repeat for Understanding Object Articulations from Interactions", IEEE International Conference on Robotics and Automation (ICRA), March 2023.BibTeX TR2023-009 PDF
- @inproceedings{Ota2023mar,
- author = {Ota, Kei and Tung, Hsiao-Yu and Smith, Kevin and Cherian, Anoop and Marks, Tim K. and Sullivan, Alan and Kanezaki, Asako and Tenenbaum, Joshua B.},
- title = {H-SAUR: Hypothesize, Simulate, Act, Update, and Repeat for Understanding Object Articulations from Interactions},
- booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
- year = 2023,
- month = mar,
- url = {https://www.merl.com/publications/TR2023-009}
- }
- , "Inverse design of two-dimensional freeform metagrating using an adversarial conditional variational autoencoder", SPIE Photonics West, January 2023.BibTeX TR2023-004 PDF
- @inproceedings{Kojima2023jan,
- author = {Kojima, Keisuke and Koike-Akino, Toshiaki and Wang, Ye and Jung Minwoo and Brand, Matthew},
- title = {Inverse design of two-dimensional freeform metagrating using an adversarial conditional variational autoencoder},
- booktitle = {SPIE Photonics West},
- year = 2023,
- month = jan,
- url = {https://www.merl.com/publications/TR2023-004}
- }
- , "Learning a Constrained Optimizer: A Primal Method", AAAI Conference on Artificial Intelligence, January 2023.BibTeX TR2023-003 PDF
- @inproceedings{Liu2023jan,
- author = {Liu, Tao and Cherian, Anoop},
- title = {Learning a Constrained Optimizer: A Primal Method},
- booktitle = {AAAI Conference on Artificial Intelligence},
- year = 2023,
- month = jan,
- url = {https://www.merl.com/publications/TR2023-003}
- }
- , "GSR: A Generalized Symbolic Regression Approach", Transactions on Machine Learning Research, January 2023.BibTeX TR2023-002 PDF
- @article{Tohme2023jan,
- author = {Tohme, Tony and Liu, Dehong and Youcef-Toumi, Kamal},
- title = {GSR: A Generalized Symbolic Regression Approach},
- journal = {Transactions on Machine Learning Research},
- year = 2023,
- month = jan,
- url = {https://www.merl.com/publications/TR2023-002}
- }
- , "STFT-Domain Neural Speech Enhancement with Very Low Algorithmic Latency", IEEE/ACM Transactions on Audio, Speech, and Language Processing, December 2022.BibTeX TR2022-166 PDF
- @article{Wang2022dec2,
- author = {Wang, Zhong-Qiu and Wichern, Gordon and Watanabe, Shinji and Le Roux, Jonathan},
- title = {STFT-Domain Neural Speech Enhancement with Very Low Algorithmic Latency},
- journal = {IEEE/ACM Transactions on Audio, Speech, and Language Processing},
- year = 2022,
- month = dec,
- url = {https://www.merl.com/publications/TR2022-166}
- }
- , "Learning with noisy labels using low-dimensional model trajectory", NeurIPS 2022 Workshop on Distribution Shifts (DistShift), December 2022.BibTeX TR2022-156 PDF
- @inproceedings{Singla2022dec,
- author = {Singla, Vasu and Aeron, Shuchin and Koike-Akino, Toshiaki and Parsons, Kieran and Brand, Matthew and Wang, Ye},
- title = {Learning with noisy labels using low-dimensional model trajectory},
- booktitle = {NeurIPS 2022 Workshop on Distribution Shifts (DistShift)},
- year = 2022,
- month = dec,
- url = {https://www.merl.com/publications/TR2022-156}
- }
- , "Are Deep Neural Networks SMARTer than Second Graders?", IEEE Conference on Computer Vision and Pattern Recognition (CVPR), March 2023.
-
Videos
-
Human Perspective Scene Understanding via Multimodal Sensing
-
[MERL Seminar Series Spring 2022] Self-Supervised Scene Representation Learning
-
[MERL Seminar Series Spring 2022] Learning Speech Representations with Multimodal Self-Supervision
-
[MERL Seminar Series 2021] Deep probabilistic regression
-
[MERL Seminar Series 2021] Learning to See by Moving: Self-supervising 3D scene representations for perception, control, and visual reasoning
-
[MERL Seminar Series 2021] Look and Listen: From Semantic to Spatial Audio-Visual Perception
-
Application of Deep Learning for Nanophotonic Device Design (Invited)
-
Machine Learning Power Amplifier
-
Scene-Aware Interaction Technology
-
-
Software Downloads
-
SOurce-free Cross-modal KnowledgE Transfer
-
Nonparametric Score Estimators
-
Instance Segmentation GAN
-
Audio Visual Scene-Graph Segmentor
-
Generalized One-class Discriminative Subspaces
-
Goal directed RL with Safety Constraints
-
Hierarchical Musical Instrument Separation
-
Generating Visual Dynamics from Sound and Context
-
Adversarially-Contrastive Optimal Transport
-
Online Feature Extractor Network
-
MotionNet
-
FoldingNet++
-
Quasi-Newton Trust Region Policy Optimization
-
Landmarks’ Location, Uncertainty, and Visibility Likelihood
-
Robust Iterative Data Estimation
-
Gradient-based Nikaido-Isoda
-
Discriminative Subspace Pooling
-
Partial Group Convolutional Neural Networks
-
