The anti-k_T algorithm is the default jet clustering algorithm at the LHC. It clusters particles based on distance measures d_{ij} = min(1/p_{Ti}^2, 1/p_{Tj}^2) * Delta R_{ij}^2 / R^2 and d_{iB} = 1/p_{Ti}^2. What property makes it preferred over earlier algorithms like the k_T algorithm?
AIt runs faster computationally
BIt produces perfectly circular, cone-like jets with hard particles at the center, while being IRC safe — soft particles at the jet boundary do not distort the jet shape, making jet calibration and comparison with theory straightforward
CIt always produces exactly two jets
DIt eliminates the need for jet energy corrections
The anti-k_T algorithm preferentially clusters soft particles with nearby hard particles (because the 1/p_T^2 weighting favors high-p_T seeds). This produces jets with regular, circular boundaries around hard cores. The k_T algorithm (which uses p_T^2 instead of 1/p_T^2) clusters soft particles first, producing irregular jet shapes. Both are IRC safe and give the same inclusive cross sections, but anti-k_T jets are experimentally easier to calibrate and correct.
Question 2 Short Answer
A jet algorithm must be 'infrared and collinear (IRC) safe' to be useful for QCD calculations. What does this requirement mean physically?
Think about your answer, then reveal below.
Model answer: IRC safety means the set of jets found by the algorithm does not change when a soft (zero-energy) particle is added to the event (infrared safety) or when a particle is split into two collinear particles sharing its momentum (collinear safety). Perturbative QCD calculations contain infrared and collinear divergences that cancel between real-emission and virtual-correction diagrams — but this cancellation works only if the observable (here, the jet definition) is insensitive to soft and collinear emissions. An IRC-unsafe algorithm would give different jet multiplicities or momenta in the presence of soft radiation, making the perturbative calculation divergent and the observable ill-defined.
The requirement of IRC safety eliminated many early cone-based jet algorithms that used seed-based iterative procedures. The sequential recombination algorithms (k_T, Cambridge/Aachen, anti-k_T) are all IRC safe by construction.
Question 3 Multiple Choice
The parameter R in jet algorithms sets the jet 'radius' in eta-phi space. At the LHC, typical choices are R = 0.4 for resolved jets and R = 0.8 or 1.0 for 'fat jets.' Why do analyses of boosted heavy particles (top quarks, W/Z/H bosons) use large-R jets?
ABecause heavy particles produce more particles in their decay
BWhen a heavy particle is produced with transverse momentum much larger than its mass, its decay products are collimated into a cone of angular size approximately 2m/p_T — a large-R jet captures the entire decay, and jet substructure techniques can then identify the decay pattern inside the single jet
CBecause large-R jets have better energy resolution
DBecause QCD background is lower for large-R jets
A top quark with p_T = 500 GeV decaying to Wb -> qqb produces three quarks separated by Delta R ~ 2*m_t/p_T ~ 0.7. A small-R jet (R=0.4) would resolve these as separate jets, while a large-R jet (R=1.0) captures them all. Jet substructure techniques (trimming, pruning, soft-drop, N-subjettiness) then analyze the internal structure of the fat jet to distinguish boosted top quarks from QCD background. This 'boosted object tagging' has become a major tool at the LHC.