no code implementations • 30 Jan 2024 • Sayeed Shafayet Chowdhury, Soumyadeep Chandra, Kaushik Roy
Through our experiments, we discover an intriguing property of DNNs where we observe that state-of-the-art convolutional neural networks, as well as vision transformers, fail to discriminate between syntactically correct and incorrect images when trained on only correct ones.
1 code implementation • 24 May 2023 • Yinghan Long, Sayeed Shafayet Chowdhury, Kaushik Roy
The segmented attention and lightweight RAF neurons ensure the efficiency of the proposed transformer.
Ranked #1 on Text Summarization on MediaSum
no code implementations • 2 Oct 2021 • Ruhi Sharmin, Sayeed Shafayet Chowdhury, Farihal Abedin, Kazi Mujibur Rahman
In this paper, we proposed a method using supervised ML in solar PV system for MPPT analysis.
1 code implementation • 1 Oct 2021 • Sayeed Shafayet Chowdhury, Nitin Rathi, Kaushik Roy
We achieve top-1 accuracy of 93. 05%, 70. 15% and 67. 71% on CIFAR-10, CIFAR-100 and ImageNet, respectively using VGG16, with just 1 timestep.
no code implementations • 26 Apr 2021 • Sayeed Shafayet Chowdhury, Isha Garg, Kaushik Roy
Moreover, they require 8-14X lesser compute energy compared to their unpruned standard deep learning counterparts.
no code implementations • ICCV 2021 • Isha Garg, Sayeed Shafayet Chowdhury, Kaushik Roy
Notably, DCT-SNN performs inference with 2-14X reduced latency compared to other state-of-the-art SNNs, while achieving comparable accuracy to their standard deep learning counterparts.
1 code implementation • 5 Oct 2020 • Isha Garg, Sayeed Shafayet Chowdhury, Kaushik Roy
Notably, DCT-SNN performs inference with 2-14X reduced latency compared to other state-of-the-art SNNs, while achieving comparable accuracy to their standard deep learning counterparts.
no code implementations • 17 Jul 2020 • Sayeed Shafayet Chowdhury, Kaji Mejbaul Islam, Rouhan Noor
Additionally, we have also tested this approach on an in-house dataset to validate its robustness.
no code implementations • 15 Jun 2020 • Sayeed Shafayet Chowdhury, Chankyu Lee, Kaushik Roy
While the leaky models have been argued as more bioplausible, a comparative analysis between models with and without leak from a purely computational point of view demands attention.
no code implementations • 5 Jun 2020 • Sayeed Shafayet Chowdhury, Kazi Mejbaul Islam, Rouhan Noor
To that effect, in this paper, a heterogeneous system is proposed which estimates the degree of abnormality of an unmanned surveillance drone, analyzing real-time image and IMU (Inertial Measurement Unit) sensor data in an unsupervised manner.