Gautam Reddy

This study examines how epistasis and pleiotropy interact in complex traits using generic models and experimental data from yeast, showing that specific mutation patterns emerge under certain conditions.

The paper shows that phenotypic heterogeneity in yeast populations is an evolutionarily stable strategy that maximizes long-term fitness under environmental changes.

This paper applies attention-based models to predict phenotype from genotype, showing improved performance in epistatic regimes and enabling transfer learning across environments.

The paper shows that manifold learning, implemented via inhibitory interneurons with IBCM plasticity, enables olfactory habituation to fluctuating backgrounds better than mean subtraction methods.

This paper proposes a Bayesian theory for temporal causal learning that explains learning phenomena through timescale invariance and integrates aspects of timing and prediction error theories.

The paper investigates how feedforward and feedback inputs to the olfactory bulb encode odor information, finding that feedback provides sparse, decorrelated representations of complex odors.

The paper shows that antagonistic interactions between odors are common in olfactory sensory neurons of freely breathing mice, helping to normalize activity and reduce saturation in complex odor mixtures.

The paper investigates how odorant receptor inhibition shapes neuronal responses to complex odors and finds that OR antagonism, inverse agonism, and partial agonism are common and combinatorial in olfactory sensory neurons.

The paper introduces a neural approximation algorithm, EXP, that outperforms existing models in approximating Bayesian prediction performance while maintaining computational efficiency through a persistent prior mechanism.

The paper models how mammalian olfactory receptor neurons process odor mixtures, showing antagonistic interactions lead to normalization of glomerular responses and improve odor discrimination.