∂u/∂t = α∇²u ∑ aₙxⁿ ∫₀^∞ f(x)dx |x − x₀| < δ ∇L(θ) x y
Tabish A.

Jack of all trades,
Master of some.

Hey there! I'm Tabish, an applied mathematician based in L'Aquila, Italy. I study systems that are hard to predict and build models that make them less so — from Kolmogorov-Arnold Networks to stochastic neural systems.

xy∂u/∂t = ν ∇²uu(x,0) = u₀(x)‖∇f(xₖ)‖ → 0L(θ)t∇²uλₙ∫ f dxθ ← θ − α∇L