Susmit Jha

Susmit Jha

Auto-generated profile, based on publicly available information

Susmit Jha

Computer Science Laboratory SRI International

Research Focus & Metrics

Academic Career Length

21Years (2006-Present)

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

Representative Publications

View All Publications
1 May 2010
Verified
Conference Paper
Open access
PDF file
SJSusmit Jha
Susmit Jha
SGSumit Gulwani
Sumit Gulwani
SSSanjit A. Seshia
Sanjit A. Seshia
ATAshish Tiwari
Ashish Tiwari

11 March 2018
Verified
Book Chapter
Closed access
PDF file
SDSouradeep Dutta
Souradeep Dutta
SJSusmit Jha
Susmit Jha
SSSriram Sankaranarayanan
Sriram Sankaranarayanan
ATAshish Tiwari
Ashish Tiwari

4 June 2011
Verified
Conference Paper
Open access
PDF file
SGSumit Gulwani
Sumit Gulwani
SJSusmit Jha
Susmit Jha
ATAshish Tiwari
Ashish Tiwari
RVRamarathnam Venkatesan
Ramarathnam Venkatesan

15 February 2017 in Acta Informatica
Verified
Journal Article
Open access
PDF file
SJSusmit Jha
Susmit Jha
SSSanjit A. Seshia
Sanjit A. Seshia
Formal synthesis is the process of generating a program satisfying a high-level formal specification. In recent times, effective formal synthesis methods have been proposed based on the use of inductive learning. We refer to this class of methods that learn programs from examples as formal inductive synthesis. In this paper, we present a theoretical framework for formal inductive synthesis. We discuss how formal inductive synthesis differs from traditional machine learning. We then describe oracle-guided inductive synthesis (OGIS), a framework that captures a family of synthesizers that operate by iteratively querying an oracle. An instance of OGIS that has had much practical impact is counterexample-guided inductive synthesis (CEGIS). We present a theoretical characterization of CEGIS for learning any program that computes a recursive language. In particular, we analyze the relative power of CEGIS variants where the types of counterexamples generated by the oracle varies. We also consider the impact of bounded versus unbounded memory available to the learning algorithm. In the special case where the universe of candidate programs is finite, we relate the speed of convergence to the notion of teaching dimension studied in machine learning theory. Altogether, the results of the paper take a first step towards a theoretical foundation for the emerging field of formal inductive synthesis.

26 May 2017 in Journal of Automated Reasoning
Verified
Journal Article
Closed access
PDF file
SJSusmit Jha
Susmit Jha
VRVasumathi Raman
Vasumathi Raman
DSDorsa Sadigh
Dorsa Sadigh
SSSanjit A. Seshia
Sanjit A. Seshia

1 December 2008
Verified
Conference Paper
Closed access
SJSusmit Jha
Susmit Jha
SJSumit Kumar Jha
Sumit Kumar Jha
In this paper, we propose a randomization based technique to verify whether a manufactured chip conforms to its design or is infected by any trojan circuit. A trojan circuit can be inserted into the design or fabrication mask by a malicious manufacturer such that it monitors for a specific rare trigger condition, and then it produces a payload error in the circuit which alters the functionality of the circuit often causing a catastrophic crash of the system where the chip was being used. Since trojans are activated by rare input patterns, they are stealthy by nature and are difficult to detect through conventional techniques of functional testing. In this paper, we propose a novel randomized approach to probabilistically compare the functionality of the implemented circuit with the design of the circuit. Using hypothesis testing, we provide quantitative guarantees when our algorithm reports that there is no trojan in the implemented circuit. This allows us to trade runtime for accuracy. The technique is sound, that is, it reports presence of a trojan only if the implemented circuit is actually infected. If our algorithm finds that the implemented circuit is infected with a trojan, it also reports a fingerprint input pattern to distinguish the implemented circuit from the design. We illustrate the effectiveness of our technique on a set of infected and benign combinational circuits.