no code implementations • 8 Feb 2024 • Md Abir Hossen, Sonam Kharade, Jason M. O'Kane, Bradley Schmerl, David Garlan, Pooyan Jamshidi
This paper proposes CURE -- a method that identifies causally relevant configuration options, enabling the optimization process to operate in a reduced search space, thereby enabling faster optimization of robot performance.
1 code implementation • 24 Aug 2023 • Saeid Ghafouri, Kamran Razavi, Mehran Salmani, Alireza Sanaee, Tania Lorido-Botran, Lin Wang, Joseph Doyle, Pooyan Jamshidi
Model variants are different versions of pre-trained models for the same deep learning task with variations in resource requirements, latency, and accuracy.
no code implementations • 2 Jun 2023 • Hamed Damirchi, Forest Agostinelli, Pooyan Jamshidi
However, a lack of structure in each module's role, and modular network-specific issues such as module collapse have restricted their usability.
1 code implementation • EuroMLSys 2023 • Mehran Salmani, Saeid Ghafouri, Alireza Sanaee, Kamran Razavi, Max Mühlhäuser, Joseph Doyle, Pooyan Jamshidi, Mohsen Sharifi
Adapting to dynamic workloads considering all the pillars of accuracy, latency, and resource cost is challenging.
1 code implementation • 5 Feb 2023 • Fatemeh Ghofrani, Mehdi Yaghouti, Pooyan Jamshidi
To advance the understanding of robust deep learning, we delve into the effects of adversarial training on self-supervised and supervised contrastive learning alongside supervised learning.
1 code implementation • 18 Jan 2023 • Md Abir Hossen, Sonam Kharade, Bradley Schmerl, Javier Cámara, Jason M. O'Kane, Ellen C. Czaplinski, Katherine A. Dzurilla, David Garlan, Pooyan Jamshidi
Finding the root cause of such faults is challenging due to the exponentially large configuration space and the dependencies between the robot's configuration settings and performance.
no code implementations • 18 Sep 2022 • Mohammadamin Abedi, Yanni Iouannou, Pooyan Jamshidi, Hadi Hemmati
The proposed solution is an automated online layer caching mechanism that allows early exiting of a large model during inference time if the cache model in one of the early exits is confident enough for final prediction.
1 code implementation • 31 May 2022 • Ali Mokhtari, Md Abir Hossen, Pooyan Jamshidi, Mohsen Amini Salehi
The challenge is to allocate user requests for different ML applications on the Heterogeneous Edge Computing Systems (HEC) with respect to both the energy and latency constraints of these systems.
no code implementations • 25 Jan 2022 • Rahul Goel, Modar Sulaiman, Kimia Noorbakhsh, Mahdi Sharifi, Rajesh Sharma, Pooyan Jamshidi, Kallol Roy
The pretrained transformer of GPT-2 is trained to generate text and then fine-tuned to classify facial images.
1 code implementation • 20 Jan 2022 • Md Shahriar Iqbal, Rahul Krishna, Mohammad Ali Javidian, Baishakhi Ray, Pooyan Jamshidi
Understanding and reasoning about the performance behavior of highly configurable systems, over a vast and variable space, is challenging.
1 code implementation • 7 Oct 2021 • Kimia Noorbakhsh, Modar Sulaiman, Mahdi Sharifi, Kallol Roy, Pooyan Jamshidi
In this paper, we present a sample efficient way of solving the symbolic tasks by first pretraining the transformer model with language translation and then fine-tuning the pretrained transformer model to solve the downstream task of symbolic mathematics.
1 code implementation • 27 Feb 2021 • Mohammad Ali Javidian, Om Pandey, Pooyan Jamshidi
To overcome this difficulty, we propose SCTL, an algorithm that avoids an exhaustive search and identifies invariant causal features across source and target domains based on Markov blanket discovery.
no code implementations • 23 Feb 2021 • Md. Musfiqur Rahman, Ayman Rasheed, Md. Mosaddek Khan, Mohammad Ali Javidian, Pooyan Jamshidi, Md. Mamun-or-Rashid
This paper proposes a generic causal structure refinement strategy that can locate the undesired relations with a small number of CI-tests, thus speeding up the algorithm for large and complex problems.
1 code implementation • 29 May 2020 • Mohammad Ali Javidian, Marco Valtorta, Pooyan Jamshidi
We provide a novel scalable and sound algorithm for Markov blanket discovery in LWF CGs and prove that the Grow-Shrink algorithm, the IAMB algorithm, and its variants are still correct for Markov blanket discovery in LWF CGs under the same assumptions as for Bayesian networks.
1 code implementation • 27 May 2020 • Mohammad Ali Javidian, Marco Valtorta, Pooyan Jamshidi
We present a PC-like algorithm that finds the structure of chain graphs under the faithfulness assumption to resolve the problem of scalability of the proposed algorithm by Studeny (1997).
no code implementations • 13 May 2020 • Yang Ren, Gregory Gay, Christian Kästner, Pooyan Jamshidi
Machine learning (ML) frameworks and the systems developed using them differ greatly from traditional frameworks.
1 code implementation • 24 Feb 2020 • Mohammad Ali Javidian, Marco Valtorta, Pooyan Jamshidi
To address the problem of learning the structure of AMP CGs from data, we show that the PC-like algorithm (Pena, 2012) is order-dependent, in the sense that the output can depend on the order in which the variables are given.
1 code implementation • 18 Jan 2020 • Md Shahriar Iqbal, Jianhai Su, Lars Kotthoff, Pooyan Jamshidi
FlexiBO weights the improvement of the hypervolume of the Pareto region by the measurement cost of each objective to balance the expense of collecting new information with the knowledge gained through objective evaluations, preventing us from performing expensive measurements for little to no gain.
3 code implementations • 2 Jan 2020 • Ying Meng, Jianhai Su, Jason O'Kane, Pooyan Jamshidi
There has been extensive research on developing defense techniques against adversarial attacks; however, they have been mainly designed for specific model families or application domains, therefore, they cannot be easily extended.
2 code implementations • 1 Nov 2019 • Rahul Krishna, Vivek Nair, Pooyan Jamshidi, Tim Menzies
To resolve these problems, we propose a novel transfer learning framework called BEETLE, which is a "bellwether"-based transfer learner that focuses on identifying and learning from the most relevant source from amongst the old data.
Software Engineering
1 code implementation • 1 Oct 2019 • Mohammad Ali Javidian, Marco Valtorta, Pooyan Jamshidi
We consider the PC-like algorithm for structure learning of MVR CGs, which is a constraint-based method proposed by Sonntag and Pe\~{n}a in [18].
1 code implementation • 2 Jul 2019 • Aaron M. Roth, Nicholay Topin, Pooyan Jamshidi, Manuela Veloso
There is a growing desire in the field of reinforcement learning (and machine learning in general) to move from black-box models toward more "interpretable AI."
1 code implementation • 4 Apr 2019 • Md Shahriar Iqbal, Lars Kotthoff, Pooyan Jamshidi
Modern deep neural network (DNN) systems are highly configurable with large a number of options that significantly affect their non-functional behavior, for example inference time and energy consumption.
1 code implementation • 10 Mar 2019 • Pooyan Jamshidi, Javier Cámara, Bradley Schmerl, Christian Kästner, David Garlan
Modern cyber-physical systems (e. g., robotics systems) are typically composed of physical and software components, the characteristics of which are likely to change over time.
1 code implementation • 26 Feb 2019 • Mohammad Ali Javidian, Pooyan Jamshidi, Marco Valtorta
We expect that the ability to carry over causal relations will enable effective performance analysis of highly-configurable systems.
3 code implementations • 11 Mar 2018 • Vivek Nair, Rahul Krishna, Tim Menzies, Pooyan Jamshidi
Using this insight, this paper proposes BEETLE, a novel bellwether based transfer learning scheme, which can identify a suitable source and use it to find near-optimal configurations of a software system.
Software Engineering
1 code implementation • 7 Sep 2017 • Pooyan Jamshidi, Norbert Siegmund, Miguel Velez, Christian Kästner, Akshay Patel, Yuvraj Agarwal
Modern software systems provide many configuration options which significantly influence their non-functional properties.
no code implementations • 19 May 2017 • Hamid Arabnejad, Claus Pahl, Pooyan Jamshidi, Giovani Estrada
A goal of cloud service management is to design self-adaptable auto-scaler to react to workload fluctuations and changing the resources assigned.
1 code implementation • 2 Jul 2015 • Pooyan Jamshidi, Amir Sharifloo, Claus Pahl, Andreas Metzger, Giovani Estrada
The benefit is that for designing cloud controllers, we do not have to rely solely on precise design-time knowledge, which may be difficult to acquire.