Research Interest
Cosmology
I’ve always been captivated by cosmology because it feels like reading the universe’s autobiography: tiny quantum ripples blooming into galaxies, supernova flashes, ancient microwave whispers, and bent starlight all weaving a story across billions of years. That wonder drives my research: I build simple, model-independent tools that let supernovae, Hubble measurements, galaxy clusters, the CMB, and gravitational lensing speak for themselves—so every new observation can surprise us, challenge our assumptions, and bring us closer to the true cosmic tale.
1. Model-Independent Frameworks
1.1 Non-parametric Regression (LOESS + SIMEX)
We applied locally weighted regression (LOESS) combined with simulation-extrapolation (SIMEX) to reconstruct the cosmic distance duality relation purely from supernova luminosity distances, radio-galaxy angular-diameter distances, and CMB temperature measurements—finding perfect consistency with expectations, all without assuming any background cosmology.
1.2 Gaussian Process Smoothing
Gaussian Processes allow me to smooth noisy Hubble-parameter data from cosmic chronometers and BAO surveys, then propagate those reconstructions into derived quantities—ensuring our conclusions don’t depend on a guessed functional form.
1.3 Hubble Phase-Space Portrait (HPSP)
With the HPSP technique, we plot the rate of change of the expansion rate against the expansion rate itself—directly from data—to pinpoint when the universe switched from slowing down to speeding up, all without invoking any dark-energy parametrization.
2. Hubble Parameter Measurements
2.1 Curvature Constraints
We can test the shape of the universe without assuming any particular cosmological model by comparing two independent distance measures. “Radial” distances come from the expansion‐rate history inferred via the ages of passively evolving galaxies, while “transverse” distances come from the angular separations in strong-lensing systems. Matching these two tells us whether space is curved or flat—and so far, the data point to a beautifully flat cosmos.
2.2 Pinpointing the Cosmic Transition
When did the universe switch from slowing down to speeding up? The Hubble Phase-Space Portrait (HPSP) method answers this by reconstructing how the expansion rate itself changes over time—straight from observed Hubble measurements, with no dark-energy fit needed. This gives a clean, model-independent estimate of the “transition redshift” marking cosmic acceleration.
3. Type Ia Supernovae as Cosmic Laboratories
3.1 Probing Distance Duality and Particle Physics
By tracking how Type Ia supernovae fade after their peak, we can test whether fundamental constants—like the strength of the weak force—have drifted over cosmic time. Matching the observed decline rates to our Gaussian-process–smoothed expansion history delivers some of the most robust, model-independent limits on any change in the Fermi coupling constant.
3.2 Non-parametric Checks on Brightness Evolution
Taking this further, we perform a joint, non-parametric analysis of supernova peak brightness evolution and the cosmic distance duality relation. The outcome? Both supernova luminosities and the standard distance scaling hold steady to better than one percent across a broad redshift range.
4. Galaxy Clusters: Weak Lensing, SZ Effect & Gas Fractions
4.1 Tightening the Graviton’s Mass Limit
Galaxy clusters act as cosmic laboratories for gravity. By comparing the total mass inferred from weak-lensing shear maps with the hot-gas pressure profiles measured via the Sunyaev–Zel’dovich effect—and tying both to our smoothed expansion history—we derive some of the strongest, model-independent upper bounds on the graviton’s mass ever reported.
4.2 Testing Distance Duality with Gas Mass Fractions
Cluster gas fractions provide a powerful, purely geometric test of photon conservation. We combine angular-diameter distances from strong-lensing systems with measurements of the ratio of gas to total mass in clusters—assuming only spatial flatness—to check the distance duality relation out to moderate redshifts. The result? No detectable violation, reinforcing the standard relation to high precision.Testing the cosmic distance duality relation with strong-lensing and gas-mass-fraction observations
5. Gravitational Lensing Applications
5.1 Strong Lensing for Distance Duality
Time-delay lenses and Einstein rings serve as independent “standard rulers.” By combining over a hundred such systems with supernova data, we achieve sub-percent tests of the distance duality relation.Probing the cosmic distance duality relation using time delay lenses6.2 Image-Separation Statistics for CurvatureThe distribution of image separations in lensed quasars provides an independent check on cosmic curvature, complementing the radial vs. transverse approach.
5.2 Weak Lensing in Cluster Mass Profiles
Weak-lensing maps of galaxy clusters underpin both the graviton-mass bounds and cluster gas-fraction tests, showcasing the versatility of this probe in a model-independent toolkit.