Author name cluster

Ashwin Srinivasan

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

11 papers

1 author row

AAAI Conference 2024 Conference Paper

Generating Novel Leads for Drug Discovery Using LLMs with Logical Feedback

Shreyas Bhat Brahmavar
Ashwin Srinivasan
Tirtharaj Dash
Sowmya Ramaswamy Krishnan
Lovekesh Vig
Arijit Roy
Raviprasad Aduri

Large Language Models (LLMs) can be used as repositories of biological and chemical information to generate pharmacological lead compounds. However, for LLMs to focus on specific drug targets typically requires experimentation with progressively more refined prompts. Results thus become dependent not just on what is known about the target, but also on what is known about the prompt- engineering. In this paper, we separate the prompt into domain-constraints that can be written in a standard logical form and a simple text-based query. We investigate whether LLMs can be guided, not by refining prompts manually, but by refining the logical component automatically, keeping the query unchanged. We describe an iterative procedure LMLF (“Language Model with Logical Feedback”) in which the constraints are progressively refined using a logical notion of generalisation. On any iteration, newly generated instances are verified against the constraint, providing "logical-feedback" for the next iteration's refinement of the constraints. We evaluate LMLF using two well-known targets (inhibition of the Janus Kinase 2; and Dopamine Receptor D2); and two different LLMs (GPT-3 and PaLM). We show that LMLF, starting with the same logical constraints and query text, can be used to guide both LLMs to generate potential leads. We find: (a) Binding affinities of LMLF-generated molecules are skewed towards higher binding affinities than those from existing baselines; (b) LMLF results in generating molecules that are skewed towards higher binding affinities than without logical feedback; (c) Assessment by a computational chemist suggests that LMLF generated compounds may be novel inhibitors. These findings suggest that LLMs with logical feedback may provide a mechanism for generating new leads without requiring the domain-specialist to acquire sophisticated skills in prompt-engineering.

PDF Details DOI

AAAI Conference 2022 Short Paper

Solving Visual Analogies Using Neural Algorithmic Reasoning (Student Abstract)

Atharv Sonwane
Gautam Shroff
Lovekesh Vig
Ashwin Srinivasan
Tirtharaj Dash

We consider a class of visual analogical reasoning problems that involve discovering the sequence of transformations by which pairs of input/output images are related, so as to analogously transform future inputs. This program synthesis task can be easily solved via symbolic search. Using a variation of the ‘neural analogical reasoning’ approach, we instead search for a sequence of elementary neural network transformations that manipulate distributed representations derived from a symbolic space, to which input images are directly encoded. We evaluate the extent to which our ‘neural reasoning’ approach generalises for images with unseen shapes and positions.

PDF Details

AAMAS Conference 2019 Conference Paper

A Reinforcement Learning Framework for Container Selection and Ship Load Sequencing in Ports

Richa Verma
Sarmimala Saikia
Harshad Khadilkar
Puneet Agarwal
Gautam Shroff
Ashwin Srinivasan

We describe a reinforcement learning (RL) framework for selecting and sequencing containers to load onto ships in ports. The goal is to minimize an approximation of the number of crane movements require to load a given ship, known as the shuffle count. It can be viewed as a version of the assignment problem in which the sequence of assignment is of importance and the task rewards are order dependent. The proposed methodology is developed specifically to be usable on ship and yard layouts of arbitrary scale, by dividing the full problem into fixed future horizon segments and through a redefinition of the action space into a binary choice framework. Using data from real-world yard and ship layouts, we show that our approach solves the single crane version of the loading problem for entire ships with better objective values than those computed using standard metaheuristics.

PDF

JMLR Journal 2019 Journal Article

Logical Explanations for Deep Relational Machines Using Relevance Information

Ashwin Srinivasan
Lovekesh Vig
Michael Bain

Our interest in this paper is in the construction of symbolic explanations for predictions made by a deep neural network. We will focus attention on deep relational machines (DRMs: a term introduced in Lodhi (2013)). A DRM is a deep network in which the input layer consists of Boolean-valued functions (features) that are defined in terms of relations provided as domain, or background, knowledge. Our DRMs differ from those in Lodhi (2013), which uses an Inductive Logic Programming (ILP) engine to first select features (we use random selections from a space of features that satisfies some approximate constraints on logical relevance and non-redundancy). But why do the DRMs predict what they do? One way of answering this was provided in recent work Ribeiro et al. (2016), by constructing readable proxies for a black-box predictor. The proxies are intended only to model the predictions of the black-box in local regions of the instance-space. But readability alone may not be enough: to be understandable, the local models must use relevant concepts in an meaningful manner. We investigate the use of a Bayes-like approach to identify logical proxies for local predictions of a DRM. As a preliminary step, we show that DRM's with our randomised propositionalization method achieve predictive performance that is comparable to the best reports in the ILP literature. Our principal results on logical explanations show: (a) Models in first-order logic can approximate the DRM's prediction closely in a small local region; and (b) Expert-provided relevance information can play the role of a prior to distinguish between logical explanations that perform equivalently on prediction alone. [abs] [ pdf ][ bib ] &copy JMLR 2019. ( edit, beta )

PDF Details

JMLR Journal 2011 Journal Article

Parameter Screening and Optimisation for ILP using Designed Experiments

Ashwin Srinivasan
Ganesh Ramakrishnan

Reports of experiments conducted with an Inductive Logic Programming system rarely describe how specific values of parameters of the system are arrived at when constructing models. Usually, no attempt is made to identify sensitive parameters, and those that are used are often given "factory-supplied" default values, or values obtained from some non-systematic exploratory analysis. The immediate consequence of this is, of course, that it is not clear if better models could have been obtained if some form of parameter selection and optimisation had been performed. Questions follow inevitably on the experiments themselves: specifically, are all algorithms being treated fairly, and is the exploratory phase sufficiently well-defined to allow the experiments to be replicated? In this paper, we investigate the use of parameter selection and optimisation techniques grouped under the study of experimental design. Screening and response surface methods determine, in turn, sensitive parameters and good values for these parameters. Screening is done here by constructing a stepwise regression model relating the utility of an ILP system's hypothesis to its input parameters, using systematic combinations of values of input parameters (technically speaking, we use a two-level fractional factorial design of the input parameters). The parameters used by the regression model are taken to be the sensitive parameters for the system for that application. We then seek an assignment of values to these sensitive parameters that maximise the utility of the ILP model. This is done using the technique of constructing a local "response surface". The parameters are then changed following the path of steepest ascent until a locally optimal value is reached. This combined use of parameter selection and response surface-driven optimisation has a long history of application in industrial engineering, and its role in ILP is demonstrated using well-known benchmarks. The results suggest that computational overheads from this preliminary phase are not substantial, and that much can be gained, both on improving system performance and on enabling controlled experimentation, by adopting well-established procedures such as the ones proposed here. [abs] [ pdf ][ bib ] &copy JMLR 2011. ( edit, beta )

PDF Details

JMLR Journal 2008 Journal Article

Incremental Identification of Qualitative Models of Biological Systems using Inductive Logic Programming

Ashwin Srinivasan
Ross D. King

The use of computational models is increasingly expected to play an important role in predicting the behaviour of biological systems. Models are being sought at different scales of biological organisation namely: sub-cellular, cellular, tissue, organ, organism and ecosystem; with a view of identifying how different components are connected together, how they are controlled and how they behave when functioning as a system. Except for very simple biological processes, system identification from first principles can be extremely difficult. This has brought into focus automated techniques for constructing models using data of system behaviour. Such techniques face three principal issues: (1) The model representation language must be rich enough to capture system behaviour; (2) The system identification technique must be powerful enough to identify substantially complex models; and (3) There may not be sufficient data to obtain both the model's structure and precise estimates of all of its parameters. In this paper, we address these issues in the following ways: (1) Models are represented in an expressive subset of first-order logic. Specifically, they are expressed as logic programs; (2) System identification is done using techniques developed in Inductive Logic Programming (ILP). This allows the identification of first-order logic models from data. Specifically, we employ an incremental approach in which increasingly complex models are constructed from simpler ones using snapshots of system behaviour; and (3) We restrict ourselves to "qualitative" models. These are non-parametric: thus, usually less data are required than for identifying parametric quantitative models. A further advantage is that the data need not be precise numerical observations (instead, they are abstractions like positive, negative, zero, increasing, decreasing and so on). We describe incremental construction of qualitative models using a simple physical system and demonstrate its application to identification of models at four scales of biological organisation, namely: (a) a predator-prey model at the ecosystem level; (b) a model for the human lung at the organ level; (c) a model for regulation of glucose by insulin in the human body at the extra-cellular level; and (d) a model for the glycolysis metabolic pathway at the cellular level. [abs] [ pdf ][ bib ] &copy JMLR 2008. ( edit, beta )

PDF Details

JMLR Journal 2003 Journal Article

An Empirical Study of the Use of Relevance Information in Inductive Logic Programming

Ashwin Srinivasan
Ross D. King
Michael E. Bain

Inductive Logic Programming (ILP) systems construct models for data using domain-specific background information. When using these systems, it is typically assumed that sufficient human expertise is at hand to rule out irrelevant background information. Such irrelevant information can, and typically does, hinder an ILP system's search for good models. Here, we provide evidence that if expertise is available that can provide a partial-ordering on sets of background predicates in terms of relevance to the analysis task, then this can be used to good effect by an ILP system. In particular, using data from biochemical domains, we investigate an incremental strategy of including sets of predicates in decreasing order of relevance. Results obtained suggest that: (a) the incremental approach identifies, in substantially less time, a model that is comparable in predictive accuracy to that obtained with all background information in place; and (b) the incremental approach using the relevance ordering performs better than one that does not (that is, one that adds sets of predicates randomly). For a practitioner concerned with use of ILP, the implication of these findings are two-fold: (1) when not all background information can be used at once (either due to limitations of the ILP system, or the nature of the domain) expert assessment of the relevance of background predicates can assist substantially in the construction of good models; and (2) good "first-cut" results can be obtained quickly by a simple exclusion of information known to be less relevant. [abs] [ pdf ][ ps.gz ][ ps ]

PDF Details

JMLR Journal 2003 Journal Article

ILP: A Short Look Back and a Longer Look Forward

David Page
Ashwin Srinivasan

Inductive logic programming (ILP) is built on a foundation laid by research in machine learning and computational logic. Armed with this strong foundation, ILP has been applied to important and interesting problems in the life sciences, engineering and the arts. This paper begins by briefly reviewing some example applications, in order to illustrate the benefits of ILP. In turn, the applications have brought into focus the need for more research into specific topics. We enumerate and elaborate five of these: (1) novel search methods; (2) incorporation of explicit probabilities; (3) incorporation of special-purpose reasoners; (4) parallel execution using commodity components; and (5) enhanced human interaction. It is our hypothesis that progress in each of these areas can greatly improve the contributions that can be made with ILP; and that, with assistance from research workers in other areas, significant progress in each of these areas is possible. [abs] [ pdf ][ ps.gz ][ ps ]

PDF Details

JMLR Journal 2003 Journal Article

Query Transformations for Improving the Efficiency of ILP Systems

Vítor Santos Costa
Ashwin Srinivasan
Rui Camacho
Hendrik Blockeel
Bart Demoen
Gerda Janssens
Jan Struyf
Henk Vandecasteele

Relatively simple transformations can speed up the execution of queries for data mining considerably. While some ILP systems use such transformations, relatively little is known about them or how they relate to each other. This paper describes a number of such transformations. Not all of them are novel, but there have been no studies comparing their efficacy. The main contributions of the paper are: (a) it clarifies the relationship between the transformations; (b) it contains an empirical study of what can be gained by applying the transformations; and (c) it provides some guidance on the kinds of problems that are likely to benefit from the transformations. [abs] [ pdf ][ ps.gz ][ ps ]

PDF Details

AIJ Journal 1996 Journal Article

Theories for mutagenicity: a study in first-order and feature-based induction

Ashwin Srinivasan
S.H. Muggleton
M.J.E. Sternberg
R.D. King

A classic problem from chemistry is used to test a conjecture that in domains for which data are most naturally represented by graphs, theories constructed with inductive logic programming (ILP) will significantly outperform those using simpler feature-based methods. One area that has long been associated with graph-based or structural representation and reasoning is organic chemistry. In this field, we consider the problem of predicting the mutagenic activity of small molecules: a property that is related to carcinogenicity, and an important consideration in developing less hazardous drugs. By providing an ILP system with progressively more structural information concerning the molecules, we compare the predictive power of the logical theories constructed against benchmarks set by regression, neural, and tree-based methods.

Details DOI

AIIM Journal 1992 Journal Article

Ripple down rules: Turning knowledge acquisition into knowledge maintenance

Paul Compton
Glenn Edwards
Byeong Kang
Leslie Lazarus
Ron Malor
Phil Preston
Ashwin Srinivasan

The most successful applications of medical expert systems seem to be in the interpretation of laboratory data. However, even in this domain, knowledge acquisition and maintenance are major problems. We have developed a knowledge acquisition technique (‘ripple down rules’) based on using knowledge only in the context in which it is acquired. The method also guides the expert to enter rules that are valid. This method trivialises knowledge acquisition so that building a pathology expert system becomes the minor daily task for the expert of correcting wrong interpretations and tuning the knowledge base to current expertise. A major expert system based on this technique, PEIRS (Pathology Expert Interpretative Reporting System), is now in use. The current limitations of the technique are that the underlying tree structure of the knowledge base may require the expert to re-enter some knowledge and that multiple diseases are handled as composite diseases.

Details DOI