5 Hits in 1.6 sec

Doubly Robust Off-Policy Actor-Critic Algorithms for Reinforcement Learning [article]

Riashat Islam, Raihan Seraj, Samin Yeasar Arnob, Doina Precup
2019 arXiv   pre-print
Often existing value-based policy gradient algorithms, however, suffer from high variance and instability particularly in continuous control tasks , Islam et al., 2017 .  ... 
arXiv:1912.05109v1 fatcat:plnttwxjrncz3fh25ig4h5q2a4

Code to Comment Translation: A Comparative Study on Model Effectiveness Errors [article]

Junayed Mahmud, Fahim Faisal, Raihan Islam Arnob, Antonios Anastasopoulos, Kevin Moran
2021 arXiv   pre-print
Automated source code summarization is a popular software engineering research topic wherein machine translation models are employed to "translate" code snippets into relevant natural language descriptions. Most evaluations of such models are conducted using automatic reference-based metrics. However, given the relatively large semantic gap between programming languages and natural language, we argue that this line of research would benefit from a qualitative investigation into the various
more » ... modes of current state-of-the-art models. Therefore, in this work, we perform both a quantitative and qualitative comparison of three recently proposed source code summarization models. In our quantitative evaluation, we compare the models based on the smoothed BLEU-4, METEOR, and ROUGE-L machine translation metrics, and in our qualitative evaluation, we perform a manual open-coding of the most common errors committed by the models when compared to ground truth captions. Our investigation reveals new insights into the relationship between metric-based performance and model prediction errors grounded in an empirically derived error taxonomy that can be used to drive future research efforts
arXiv:2106.08415v1 fatcat:zouqwm43djdm7mxbm6h4gpkqgu

Table of contents

2021 2021 2nd International Conference on Robotics, Electrical and Signal Processing Techniques (ICREST)  
Mohsin Sarker Raihan, M. Raihan and Meer Shadman Saeed Samia Yasmin, Md.  ...  Islam, Tahmidul Islam, Md. Raiyan Md.  ... 
doi:10.1109/icrest51555.2021.9331045 fatcat:uuunomvydvfjxd32gfpgxfuwuu

ICCIT 2020 Conference Proceedings [Front matter]

2020 2020 23rd International Conference on Computer and Information Technology (ICCIT)  
Rabiul Islam.  ...  Rabiul Islam.  ...  Finally, we ended the research by comparing and analyzing all the achieved results. 331 Soumik Sarker, Arnob Kumar Saha and Md Sadek Ferdous.  ... 
doi:10.1109/iccit51783.2020.9392749 fatcat:pz3hf7rsmzbjpe6hxjlu5tmrfq

SLAW: Scaled Loss Approximate Weighting for Efficient Multi-Task Learning [article]

Michael Crawshaw, Jana Košecká
2021 arXiv   pre-print
Stein, Abhishek Paudel, Raihan Islam Arnob, Roshan Dhakal, and Sulabh Shrestha for feedback on an early draft, and Daniel Brogan for feedback on the formatting of the proof of Theorem 1.  ... 
arXiv:2109.08218v1 fatcat:bpwlymzckzab7jsrjm7vjitgpa