Hamiltonian Bypass in Graph Theory

• Hamiltonian Bypass is a spanning subdigraph derived from a Hamiltonian cycle by reversing exactly one arc, ensuring the initial vertex dominates the terminal vertex.
• It refines Hamiltonicity by imposing specific degree conditions—such as the Bang-Jensen–Gutin–Li and Manoussakis’ conditions—to guarantee the existence of the bypass.
• Beyond graph theory, the concept extends to optimization and quantum control, where modified Hamiltonians help bypass obstacles in energy landscapes and adiabatic processes.

In graph theory, a Hamiltonian bypass is a spanning subdigraph obtained from a Hamiltonian cycle by reversing exactly one arc; equivalently, it may be described as a Hamiltonian path whose initial vertex dominates the terminal vertex (Darbinyan et al., 2014, Darbinyan, 21 Jul 2025). The notion refines Hamiltonicity by asking not merely for a directed cycle through all vertices, but for a Hamiltonian structure with a prescribed one-arc reversal. Within the supplied literature, the term also appears in broader Hamiltonian settings as an interpretive label for constructions that “bypass” obstacles in an energy landscape or an adiabatic evolution, although its canonical technical meaning is the digraph-theoretic one (Apte et al., 2022, Karzig et al., 2015, Holtzman et al., 2024).

1. Definition, notation, and equivalent formulations

For a digraph $D$, a Hamiltonian cycle is a directed cycle that visits every vertex exactly once. If

$C=x_1x_2\cdots x_nx_1$

is a Hamiltonian cycle, then a Hamiltonian bypass is obtained by reversing exactly one arc, for example replacing $x_nx_1$ by $x_1x_n$ while keeping the remaining cycle arcs (Darbinyan et al., 2014). The paper "On Hamiltonian Bypasses in one Class of Hamiltonian Digraphs" denotes such a bypass by

$D(n;2)=[x_1\to x_n;\;x_1x_2\cdots x_n],$

that is, the subdigraph consisting of the $(n-1)$-arc directed path $x_1x_2\cdots x_n$ together with the reversed arc $x_1\to x_n$ (Darbinyan et al., 2014).

A later paper adopts an equivalent path-based formulation: a Hamiltonian bypass is a Hamiltonian path $P=x_1x_2\cdots x_p$ such that the initial vertex $x_1$ dominates the terminal vertex $C=x_1x_2\cdots x_nx_1$0; equivalently, $C=x_1x_2\cdots x_nx_1$1 together with $C=x_1x_2\cdots x_nx_1$2 forms a Hamiltonian cycle (Darbinyan, 21 Jul 2025). These formulations are compatible: the reversed-arc viewpoint emphasizes derivation from a Hamiltonian cycle, while the path viewpoint emphasizes endpoint domination.

The standard degree notation in this literature is

$C=x_1x_2\cdots x_nx_1$3

with minimum out-degree $C=x_1x_2\cdots x_nx_1$4, minimum in-degree $C=x_1x_2\cdots x_nx_1$5, and minimum semi-degree $C=x_1x_2\cdots x_nx_1$6 (Darbinyan et al., 2014). Two vertices are non-adjacent if neither arc between them is present, and a common in-neighbour of $C=x_1x_2\cdots x_nx_1$7 is a vertex $C=x_1x_2\cdots x_nx_1$8 such that $C=x_1x_2\cdots x_nx_1$9 (Darbinyan et al., 2014).

The significance of the concept lies in its placement between Hamiltonian cycles and more specialized spanning configurations. The supplied graph-theoretic papers consistently treat bypass existence as a stronger structural guarantee than Hamiltonicity alone (Darbinyan et al., 2014, Darbinyan, 2014, Darbinyan, 21 Jul 2025).

2. Degree conditions with common in-neighbours

A central line of work studies Hamiltonian bypasses under the Bang-Jensen–Gutin–Li condition

$x_nx_1$0

for every pair of non-adjacent vertices $x_nx_1$1 with a common in-neighbour (Darbinyan et al., 2014). Bang-Jensen, Gutin, and Li proved that a strong digraph satisfying this condition is Hamiltonian (Darbinyan et al., 2014). Darbinyan and Karapetyan then showed that if, in addition, $x_nx_1$2, then either the digraph contains a pre-Hamiltonian cycle of length $x_nx_1$3, or $x_nx_1$4 is even and the digraph is isomorphic to the symmetric complete bipartite digraph $x_nx_1$5 or that digraph minus one arc (Darbinyan et al., 2014).

The main contribution of "On Hamiltonian Bypasses in one Class of Hamiltonian Digraphs" is a bypass strengthening of those earlier Hamiltonicity and long-cycle results. Its Theorem 12 states that if $x_nx_1$6 is a strong digraph of order $x_nx_1$7, satisfies the same Bang-Jensen–Gutin–Li condition, has minimum out-degree at least two, and minimum in-degree at least three, then $x_nx_1$8 contains a Hamiltonian bypass (Darbinyan et al., 2014). The theorem therefore upgrades a Hamiltonian existence statement to a specific spanning subdigraph with one reversed arc.

The proof is organized around a pre-Hamiltonian cycle $x_nx_1$9 of length $x_1x_n$0 and the unique vertex $x_1x_n$1 outside $x_1x_n$2 (Darbinyan et al., 2014). Under the no-bypass assumption, Lemma 4 imposes stringent adjacency restrictions: $x_1x_n$3 cannot be adjacent with both members of any consecutive pair on $x_1x_n$4, forcing bounds such as $x_1x_n$5 and $x_1x_n$6 (Darbinyan et al., 2014). Lemmas 6 through 9 show that certain local configurations involving $x_1x_n$7 and consecutive vertices of $x_1x_n$8 immediately yield a bypass or a “good cycle,” with the minimum in-degree and out-degree assumptions used explicitly (Darbinyan et al., 2014). Lemmas 10 and 11 exclude shortcut arcs of the form $x_1x_n$9 in a counterexample without a bypass, because such arcs lead to contradictions with the degree condition (Darbinyan et al., 2014).

The argument then proceeds by induction on a block of consecutive vertices of $D(n;2)=[x_1\to x_n;\;x_1x_2\cdots x_n],$0 that are non-adjacent to $D(n;2)=[x_1\to x_n;\;x_1x_2\cdots x_n],$1, repeatedly applying partner-arc and multi-insertion lemmas to build a spanning path whose concatenation with a suitable arc through $D(n;2)=[x_1\to x_n;\;x_1x_2\cdots x_n],$2 forms a Hamiltonian bypass (Darbinyan et al., 2014). Even the extremal symmetric bipartite case is handled: although a pre-Hamiltonian cycle may fail there when $D(n;2)=[x_1\to x_n;\;x_1x_2\cdots x_n],$3 is odd, the paper states that it is still straightforward to check that a Hamiltonian bypass exists under $D(n;2)=[x_1\to x_n;\;x_1x_2\cdots x_n],$4 and $D(n;2)=[x_1\to x_n;\;x_1x_2\cdots x_n],$5 (Darbinyan et al., 2014).

The degree assumptions are not merely cosmetic. The paper gives examples $D(n;2)=[x_1\to x_n;\;x_1x_2\cdots x_n],$6 and $D(n;2)=[x_1\to x_n;\;x_1x_2\cdots x_n],$7 in which $D(n;2)=[x_1\to x_n;\;x_1x_2\cdots x_n],$8, the Bang-Jensen–Gutin–Li condition holds vacuously because there is no dominated pair of non-adjacent vertices, and yet no Hamiltonian bypass exists (Darbinyan et al., 2014). The authors conjecture that the minimum in-degree bound can be lowered from $D(n;2)=[x_1\to x_n;\;x_1x_2\cdots x_n],$9 to $(n-1)$0, but they do not provide a counterexample for $(n-1)$1 (Darbinyan et al., 2014).

3. Manoussakis’ condition and the exceptional tournament $(n-1)$2

A second major sufficient condition is Manoussakis’ Condition $(n-1)$3. For every triple $(n-1)$4 with $(n-1)$5 and $(n-1)$6 non-adjacent, the condition requires:

• if there is no arc $(n-1)$7, then

$(n-1)$8

• if there is no arc $(n-1)$9, then

$x_1x_2\cdots x_n$0

(Darbinyan, 2014).

Manoussakis proved that every strong digraph satisfying $x_1x_2\cdots x_n$1 is Hamiltonian (Darbinyan, 2014). Darbinyan had earlier shown that such a digraph either contains a pre-Hamiltonian cycle of length $x_1x_2\cdots x_n$2, or, when $x_1x_2\cdots x_n$3 is even, is isomorphic to the complete bipartite digraph with partite sets of size $x_1x_2\cdots x_n$4 (Darbinyan, 2014).

The paper "On Hamiltonian Bypasses in Digraphs with the Condition of Y. Manoussakis" strengthens this to a bypass theorem: every strongly connected digraph of order $x_1x_2\cdots x_n$5 satisfying Condition $x_1x_2\cdots x_n$6 contains a Hamiltonian bypass, unless it is isomorphic to a specific tournament $x_1x_2\cdots x_n$7 of order $x_1x_2\cdots x_n$8 (Darbinyan, 2014). The exceptional tournament has vertex set $x_1x_2\cdots x_n$9, a directed $x_1\to x_n$0-cycle $x_1\to x_n$1, the chords $x_1\to x_n$2 and $x_1\to x_n$3, and the additional arcs $x_1\to x_n$4, $x_1\to x_n$5, $x_1\to x_n$6 (Darbinyan, 2014). According to the supplied details, this tournament satisfies $x_1\to x_n$7 but does not contain a Hamiltonian bypass (Darbinyan, 2014).

The proof again begins with a pre-Hamiltonian cycle $x_1\to x_n$8 and a unique off-cycle vertex $x_1\to x_n$9 (Darbinyan, 2014). Under the assumption that no bypass exists, Lemma 7 imposes bypass-avoidance constraints: for every consecutive pair on $P=x_1x_2\cdots x_p$0, at most one of the two arcs from $P=x_1x_2\cdots x_p$1 to the pair and at most one to $P=x_1x_2\cdots x_p$2 from the pair can be present; moreover $P=x_1x_2\cdots x_p$3, $P=x_1x_2\cdots x_p$4, and $P=x_1x_2\cdots x_p$5 (Darbinyan, 2014). If both $P=x_1x_2\cdots x_p$6 and $P=x_1x_2\cdots x_p$7 occur, then $P=x_1x_2\cdots x_p$8 has no arc to any other $P=x_1x_2\cdots x_p$9 (Darbinyan, 2014).

The paper combines these restrictions with path insertion lemmas and a degree trade-off lemma of Manoussakis type. One formulation given in the supplied data is that if $x_1$0 has two distinct non-adjacent pairs $x_1$1 and $x_1$2 and one of those pairs has a relatively small degree sum, then the other must have a correspondingly larger one; in particular, if $x_1$3, then $x_1$4 (Darbinyan, 2014). Through a detailed case analysis of all possible adjacency patterns between $x_1$5 and the cycle $x_1$6, the no-bypass assumption is shown to force either a contradiction with $x_1$7 or the small exceptional structure $x_1$8 (Darbinyan, 2014).

This theorem places bypass existence strictly above pre-Hamiltonian cycle existence in the $x_1$9 class. The complete bipartite digraph remains extremal for the earlier pre-Hamiltonian result, but is not exceptional for bypasses: the paper states that it still contains a Hamiltonian bypass (Darbinyan, 2014).

4. One-exception high-degree digraphs and balanced bipartite digraphs

The 2025 paper "On Hamiltonian bypasses in digraphs and bipartite digraphs" studies a different high-degree framework (Darbinyan, 21 Jul 2025). Let $C=x_1x_2\cdots x_nx_1$00 be a $C=x_1x_2\cdots x_nx_1$01-strong digraph of order $C=x_1x_2\cdots x_nx_1$02, and fix a vertex $C=x_1x_2\cdots x_nx_1$03 such that every vertex other than $C=x_1x_2\cdots x_nx_1$04 has degree at least $C=x_1x_2\cdots x_nx_1$05:

$C=x_1x_2\cdots x_nx_1$06

(Darbinyan, 21 Jul 2025). The paper introduces a conjecture asking for the smallest integer $C=x_1x_2\cdots x_nx_1$07 such that $C=x_1x_2\cdots x_nx_1$08 forces a Hamiltonian bypass, and proves that one may take $C=x_1x_2\cdots x_nx_1$09 (Darbinyan, 21 Jul 2025).

Its first principal theorem states: if either $C=x_1x_2\cdots x_nx_1$10 is Hamiltonian or

$C=x_1x_2\cdots x_nx_1$11

then $C=x_1x_2\cdots x_nx_1$12 contains a Hamiltonian bypass (Darbinyan, 21 Jul 2025). This improves an earlier threshold $C=x_1x_2\cdots x_nx_1$13 in the same one-exception large-degree setting (Darbinyan, 21 Jul 2025). The proof uses a long cycle $C=x_1x_2\cdots x_nx_1$14 of length $C=x_1x_2\cdots x_nx_1$15 containing all high-degree vertices, then derives adjacency inequalities around $C=x_1x_2\cdots x_nx_1$16 under the no-bypass assumption:

$C=x_1x_2\cdots x_nx_1$17

$C=x_1x_2\cdots x_nx_1$18

for all appropriate indices (Darbinyan, 21 Jul 2025). Summing yields

$C=x_1x_2\cdots x_nx_1$19

hence

$C=x_1x_2\cdots x_nx_1$20

so $C=x_1x_2\cdots x_nx_1$21, contradicting the hypothesis (Darbinyan, 21 Jul 2025).

The same paper also proves a bipartite theorem. Let $C=x_1x_2\cdots x_nx_1$22 be a strong balanced bipartite digraph of order $C=x_1x_2\cdots x_nx_1$23 with partite sets $C=x_1x_2\cdots x_nx_1$24 and $C=x_1x_2\cdots x_nx_1$25. If for every pair $C=x_1x_2\cdots x_nx_1$26, $C=x_1x_2\cdots x_nx_1$27 with $C=x_1x_2\cdots x_nx_1$28,

$C=x_1x_2\cdots x_nx_1$29

then $C=x_1x_2\cdots x_nx_1$30 contains a Hamiltonian bypass (Darbinyan, 21 Jul 2025). The lower bound $C=x_1x_2\cdots x_nx_1$31 is sharp: the digraph $C=x_1x_2\cdots x_nx_1$32 of order $C=x_1x_2\cdots x_nx_1$33 satisfies the weaker condition $C=x_1x_2\cdots x_nx_1$34 for every missing cross-part arc and has no Hamiltonian bypass (Darbinyan, 21 Jul 2025). By contrast, the small exceptional digraph $C=x_1x_2\cdots x_nx_1$35 satisfies the sharp condition $C=x_1x_2\cdots x_nx_1$36, is not Hamiltonian, but still contains a Hamiltonian bypass (Darbinyan, 21 Jul 2025).

These results link bypasses to classical high-degree Hamiltonicity theory. The paper recalls, for example, that every strong digraph of order $C=x_1x_2\cdots x_nx_1$37 with minimum degree at least $C=x_1x_2\cdots x_nx_1$38 contains a $C=x_1x_2\cdots x_nx_1$39, described there as a directed Hamiltonian bypass structure (Darbinyan, 21 Jul 2025). It also formulates broader conjectures, including the global conjecture that if $C=x_1x_2\cdots x_nx_1$40 vertices of a $C=x_1x_2\cdots x_nx_1$41-strong digraph of order $C=x_1x_2\cdots x_nx_1$42 have degrees at least $C=x_1x_2\cdots x_nx_1$43, then the digraph contains a Hamiltonian bypass with no degree requirement on the remaining vertex (Darbinyan, 21 Jul 2025).

5. Structural techniques, extremal examples, and open questions

Across the digraph literature, bypass proofs rely on a recurring toolkit: cycle-extension lemmas, Bondy–Thomassen-style path insertion, and the multi-insertion lemma of Bang-Jensen–Gutin (Darbinyan et al., 2014, Darbinyan, 2014, Darbinyan, 21 Jul 2025). The operative idea is to locate “partners,” namely arcs $C=x_1x_2\cdots x_nx_1$44 on a long path or cycle that permit insertion of an external vertex or short path through arcs $C=x_1x_2\cdots x_nx_1$45 and $C=x_1x_2\cdots x_nx_1$46 (Darbinyan et al., 2014). When sufficiently many such partners exist, repeated insertion yields a spanning path that closes to a bypass.

A second common theme is the analysis of a unique off-cycle vertex relative to a pre-Hamiltonian cycle. In both the Bang-Jensen–Gutin–Li and Manoussakis settings, a no-bypass assumption forces sparse and highly patterned adjacency between the outside vertex and consecutive vertices of the cycle (Darbinyan et al., 2014, Darbinyan, 2014). Those local restrictions then interact with global degree lower bounds to produce contradictions.

Extremal and exceptional constructions serve two functions: they show that bypass results are sharper than Hamiltonicity alone, and they delimit necessary hypotheses. The examples $C=x_1x_2\cdots x_nx_1$47 and $C=x_1x_2\cdots x_nx_1$48 in (Darbinyan et al., 2014) show that if the minimum semi-degree drops to $C=x_1x_2\cdots x_nx_1$49, then the Bang-Jensen–Gutin–Li condition may hold vacuously while no Hamiltonian bypass exists. The tournament $C=x_1x_2\cdots x_nx_1$50 in (Darbinyan, 2014) shows that even Manoussakis’ strong triple condition does not force a bypass in every strong digraph. The bipartite example $C=x_1x_2\cdots x_nx_1$51 in (Darbinyan, 21 Jul 2025) shows that the balanced bipartite threshold $C=x_1x_2\cdots x_nx_1$52 cannot be reduced to $C=x_1x_2\cdots x_nx_1$53.

Several open problems remain explicit in the supplied data. One is whether the theorem of (Darbinyan et al., 2014) remains valid under $C=x_1x_2\cdots x_nx_1$54 and $C=x_1x_2\cdots x_nx_1$55, rather than $C=x_1x_2\cdots x_nx_1$56. Another is the characterization problem posed in (Darbinyan, 2014): characterize the digraphs satisfying Bang-Jensen–Gutin–Li or Bang-Jensen–Guo–Yeo Hamiltonicity conditions that do not contain a Hamiltonian bypass. The 2025 paper adds conjectures on one-exception high-degree digraphs, four-vertex degree-sum conditions, Woodall-type thresholds for $C=x_1x_2\cdots x_nx_1$57, and a characterization problem for Hamiltonian balanced bipartite digraphs satisfying the weaker condition $C=x_1x_2\cdots x_nx_1$58 (Darbinyan, 21 Jul 2025).

A plausible implication is that the bypass problem has become a fine-grained refinement of degree-based Hamiltonicity theory. The supplied results consistently move from Hamiltonian existence to a more rigid spanning configuration, and the exceptions that remain are small or highly structured.

6. Broader non-graph-theoretic uses of the expression

Outside graph theory, the supplied data uses “Hamiltonian bypass” more loosely as a descriptive label for modifying a Hamiltonian so as to evade an obstruction in optimization or adiabatic evolution.

In non-convex optimization, "Non-Convex Optimization by Hamiltonian Alternation" introduces an alternate Hamiltonian

$C=x_1x_2\cdots x_nx_1$59

with the explicit choice

$C=x_1x_2\cdots x_nx_1$60

so that minima of $C=x_1x_2\cdots x_nx_1$61 below an energy threshold $C=x_1x_2\cdots x_nx_1$62 remain minima of $C=x_1x_2\cdots x_nx_1$63, while local minima of $C=x_1x_2\cdots x_nx_1$64 above $C=x_1x_2\cdots x_nx_1$65 become local maxima of $C=x_1x_2\cdots x_nx_1$66 because

$C=x_1x_2\cdots x_nx_1$67

changes sign at $C=x_1x_2\cdots x_nx_1$68 (Apte et al., 2022). The paper describes Hamiltonian alternation as repeatedly minimizing $C=x_1x_2\cdots x_nx_1$69 and $C=x_1x_2\cdots x_nx_1$70 in sequence, thereby allowing the optimizer to escape shallow minima. The supplied data states that this can be viewed as a “Hamiltonian bypass” mechanism and that the terminology “corresponds precisely” to that interpretation (Apte et al., 2022). In Sherrington–Kirkpatrick spin-glass experiments with $C=x_1x_2\cdots x_nx_1$71 and $C=x_1x_2\cdots x_nx_1$72, the paper used greedy single-spin flips and set $C=x_1x_2\cdots x_nx_1$73 to one percent less than the lowest energy attained so far; for $C=x_1x_2\cdots x_nx_1$74, higher alternation ratios yielded lower energies than pure multi-start, whereas for $C=x_1x_2\cdots x_nx_1$75 alternation offered little advantage (Apte et al., 2022).

In quantum control, the supplied description of "Shortcuts to nonabelian braiding" explicitly frames counterdiabatic driving as a “Hamiltonian bypass” for finite-time nonabelian braiding (Karzig et al., 2015). The original time-dependent Hamiltonian $C=x_1x_2\cdots x_nx_1$76 is augmented by a counterdiabatic term

$C=x_1x_2\cdots x_nx_1$77

equivalently

$C=x_1x_2\cdots x_nx_1$78

so that the exact finite-time evolution reproduces the ideal adiabatic Wilczek–Zee holonomy within a degenerate manifold (Karzig et al., 2015). For the Majorana Y-junction model

$C=x_1x_2\cdots x_nx_1$79

the resulting correction simplifies to

$C=x_1x_2\cdots x_nx_1$80

(Karzig et al., 2015). Exact implementation removes leakage and phase errors identically for any protocol duration, while approximate implementation with $C=x_1x_2\cdots x_nx_1$81 or $C=x_1x_2\cdots x_nx_1$82 still suppresses errors substantially, with residual errors scaling approximately as $C=x_1x_2\cdots x_nx_1$83 (Karzig et al., 2015).

A related control-theoretic usage appears in "Shortcuts to adiabaticity across a separatrix," where the supplied data states that the method presents a Hamiltonian bypass across a classical separatrix (Holtzman et al., 2024). There the total Hamiltonian is

$C=x_1x_2\cdots x_nx_1$84

with the velocity field chosen as

$C=x_1x_2\cdots x_nx_1$85

to transport every point on an initial energy shell to a target shell of the same enclosed Liouville volume, even across a separatrix where ordinary adiabatic transport fails (Holtzman et al., 2024). The schedule is designed so that $C=x_1x_2\cdots x_nx_1$86 at the crossing, keeping the product $C=x_1x_2\cdots x_nx_1$87 finite despite the divergence of $C=x_1x_2\cdots x_nx_1$88 at the fixed point (Holtzman et al., 2024). In the double-well erasure application, the work invested in the Hamiltonian stages is

$C=x_1x_2\cdots x_nx_1$89

and both the energy cost and the fidelity are independent of protocol duration $C=x_1x_2\cdots x_nx_1$90 (Holtzman et al., 2024).

Finally, "Optimal shortcut approach based on an easily obtained intermediate Hamiltonian" is summarized in the supplied data as a “Hamiltonian Bypass” method for shortcut design (Chen et al., 2017). Instead of adding the difficult counterdiabatic Hamiltonian of the original system, the method constructs an intermediate Hamiltonian

$C=x_1x_2\cdots x_nx_1$91

that preserves the operator structure of $C=x_1x_2\cdots x_nx_1$92 while canceling problematic matrix elements, and then adds the counterdiabatic term of $C=x_1x_2\cdots x_nx_1$93 to obtain the final transitionless Hamiltonian

$C=x_1x_2\cdots x_nx_1$94

(Chen et al., 2017). In the off-resonant $C=x_1x_2\cdots x_nx_1$95-system example, this removes the direct $C=x_1x_2\cdots x_nx_1$96 coupling from the implementable Hamiltonian, leaving only modified pump, Stokes, and detuning controls (Chen et al., 2017).

Taken together, these non-graph-theoretic uses suggest a broader metaphor: a Hamiltonian is altered so that an otherwise obstructed path—through a local-minimum basin, a diabatic braiding process, or a separatrix crossing—can be traversed by an auxiliary or deformed dynamics. That broader usage, however, remains distinct from the established graph-theoretic definition of a Hamiltonian bypass as a one-arc reversal of a Hamiltonian cycle.