Indranil Saha

Indranil Saha

Auto-generated profile, based on publicly available information

Indranil Saha

Department of Computer Science and Engineering, Indian Institute of Technology Kanpur, Kanpur, India

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
9 August 2022 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
SMSamvid Mistry
Samvid Mistry
ISIndranil Saha
Indranil Saha
SBSwarnendu Biswas
Swarnendu Biswas
Quantized Deep Neural Networks (DNNs) have the potential to find wide applications in safety-critical cyber-physical systems implemented on processors supporting only integer arithmetic. The significant challenge therein is to ensure the correctness of the operation of the network with its approximated computation. To address this verification challenge formally, we present a methodology to encode the verification problem into a Mixed-Integer Linear Programming (MILP) problem. Our encoding is based on the bit-precise semantics of quantized neural networks, which ensures the soundness of our method. We implement our verification methodology using the Gurobi MILP solver and evaluate it on several widely used DNN benchmarks. We compare our method with state-of-the-art bit-vector encodings, which are outperformed by our MILP-based verification methodology by an order of magnitude in most cases. These experimental results establish our MILP-based verification technique as a powerful tool for developing formally verified safety-critical systems with quantized DNNs as a component.

1 November 2020 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
NSNikhil Kumar Singh
Nikhil Kumar Singh
ISIndranil Saha
Indranil Saha

6 May 2025
Verified
Conference Paper
Open access
AAAman Aryan
Aman Aryan
MMManan Modi
Manan Modi
ISIndranil Saha
Indranil Saha
RMRupak Majumdar
Rupak Majumdar
SMSwarup Mohalik
Swarup Mohalik

13 December 2023
Verified
Conference Paper
Open access
TDThao Dang
Thao Dang
ADAlexandre Donzé
Alexandre Donzé
IHInzemamul Haque
Inzemamul Haque
NKNikolaos Kekatos
Nikolaos Kekatos
ISIndranil Saha
Indranil Saha
We present a novel method for imitation learning for control requirements expressed using Signal Temporal Logic (STL). More concretely we focus on the problem of training a neural network to imitate a complex controller. The learning process is guided by efficient data aggregation based on counter-examples and a coverage measure. Moreover, we introduce a method to evaluate the performance of the learned controller via parameterization and parameter estimation of the STL requirements. We demonstrate our approach with a flying robot case study.

9 May 2023
Verified
Conference Paper
Open access
AAArvind Adimoolam
Arvind Adimoolam
ISIndranil Saha
Indranil Saha
TDThao Dang
Thao Dang
Self-triggered controllers have the potential to improve the state-of-the-art of Cyber-Physical Systems (CPSs) by enhancing the performance of the underlying closed-loop control systems. However, a major concern in deploying a self-triggered controller in a safety-critical CPS is that the stabilizing self-triggered controller may not always guarantee the satisfaction of the safety constraints. We propose a self-triggered control scheme that deals with the safe scheduling of control tasks for uncertain continuous-time linear systems. We derive a computationally efficient scheduling function that computes an upper bound on the next sampling period as a function of the current state in the presence of additive disturbance. To reduce the computational complexity of online reachability analysis and increase accuracy, we compute a large sequence of reachable sets offline and use these precomputed sets to derive a low-complexity online scheduling function that computes sufficiently large bounds in real time. We evaluate our algorithm on three high-dimensional benchmark control systems, where two of the examples have a twelve-dimensional joint state plus feedback input. Experimental results demonstrate that our self-triggered control algorithm guarantees the safety of the closed-loop control system through negligible online computation, establishing the feasibility of its practical implementation.

12 May 2025 in ACM Transactions on Embedded Computing Systems
Verified
Journal Article
Open access
JBJens Brandt
Jens Brandt
ISIndranil Saha
Indranil Saha
LZLijun Zhang
Lijun Zhang