Let $a_i,c_i$ be nonzero real constants with $a_i{c}_i>0$. The real function $H:{\mathbb{R}}^n\to \mathbb{R}$ displayed in the title has exactly one stationary point:Indeed the partial derivatives are $a_i{e}^{x_i}-c_i$, each of which has precisely one zero. The Hessian matrix at that point is $\mathrm{diag}(c_1,\dots ,c_n)$.

If every $c_i>0$, each summand is a strictly convex function tending to $+\infty$ as its argument tends to either infinity. Consequently $H$ has a unique global minimum. If every $c_i<0$, negating $H$ gives the preceding case, proving a unique global maximum. If both signs occur, restricting to two coordinate directions of opposite signs gives a saddle point of a scalar function. No degenerate Hessian matrix is possible under the nonzero assumptions.

The zero level set of $F_R$ is the union of the coordinate axes and the circle of radius $R$. The gradient isIts nine stationary points are the origin, the four axis points at distance $R$, and the four points with $x^2=y^2=R^2/4$. The factored gradient proves completeness: a zero coordinate gives the five axis cases; two nonzero coordinates give a linear system in $x^2,y^2$.

The Hessian matrix has diagonal entries $6xy$ and off-diagonal entries $3(x^2+y^2)-R^2$. At the origin its eigenvalues are $\pm R^2$; at the four unit-direction axis points they are $\pm 2R^2$. These are saddle points of a scalar function. At the two interior points with equal coordinate signs, the eigenvalues are $R^2,2R^2$, giving strict local minima of value $-R^4/8$. Opposite coordinate signs give the negative eigenvalues and strict local maxima of value $R^4/8$.

On the zero circle, $\nabla F_R=2xy(x,y{)}^T$. Hence its direction is radial outward for $xy>0$, inward for $xy<0$, and zero at the four axis crossings. This simple geometric sign rule links the zero level set to the gradient sketch without requiring numerical differentiation.

The stationary points of this polynomial are the origin and $(a,a)$, $(-a,-a)$. Its Hessian matrix at the origin has eigenvalues $\pm 4a^2$, so the origin is a saddle point. At each of the other points the eigenvalues are $8a^2$ and $16a^2$, giving strict local minima. The identityshows that they are also the only global minima, with value $-2a^4$.

For $-2a^4<c<0$, the level set $g=c$ has two closed components, one surrounding each minimum. The zero level set has two lobes meeting at the origin with the coordinate axes as tangent directions. For $c>0$, there is one closed outer component. These statements also follow in polar coordinates from the quadratic equation in $r^2$, together with the absence of other stationary points.

Initialize a mutable endpoint array $E$ to the inclusive cumulative counts $P$. Scan input array indices $i=n-1,n-2,\dots ,0$. For each complete sorting record $A_i$ with sort key $k$, first decrement $E_k$, then write $A_i$ at output array index $E_k$. This is equivalent to writing at the old endpoint minus one and then decrementing the endpoint.^[ref]

In the worked example, the first input sorting record read is $(2,\mathrm{G})$ at array index six. Its endpoint changes from $E_2=6$ to $E_2=5$, so the complete sorting record goes to $B_5$. Next, $(3,\mathrm{F})$ changes $E_3$ from seven to six and goes to $B_6$. Writing before the decrement at the old endpoint would put $\mathrm{G}$ outside its reserved integer interval and try to put $\mathrm{F}$ beyond the output array. All seven writes are:Each endpoint changes only when a sorting record with its own sort key is placed. Consequently, $E_2$ stays at five through the placements of $\mathrm{F},\mathrm{E},\mathrm{D}$ before dropping to four for $\mathrm{C}$.

Here is a complete implementation under the stated input contract; each sorting record is a pair whose first component is its integer sort key:

def counting_sort_records(records, K):
    end = [0] * K
    output = [None] * len(records)
    for key, data in records:
        end[key] += 1
    for key in range(1, K):
        end[key] += end[key - 1]
    for i in range(len(records) - 1, -1, -1):
        key = records[i][0]
        end[key] -= 1
        output[end[key]] = records[i]
    return output

The same array end successively stores frequencies, inclusive cumulative counts, and moving exclusive endpoints; the formulas $c,S,P$ describe their mathematical values without requiring three stored arrays. The write copies the complete sorting record, including its associated data. In particular, two occurrences of an identical pair still cause two separate writes.

For a short algorithm correctness proof, let $r_k$ be the number of already placed sorting records with sort key $k$. A loop invariant isAll integer intervals here mean output array indices. Initially $r_k=0$, so each whole reserved integer interval is unfilled. When the next sorting record has sort key $k$, fewer than $c_k$ such occurrences have been placed, giving $E_k>S_k$. Decrementing selects a fresh slot inside its reserved integer interval. At termination, $r_k=c_k$ and $E_k=S_k$: every output slot has been filled exactly once with a complete input occurrence. The disjoint, increasing sort key ranges establish sortedness, while the fresh-slot property establishes sorting permutation preservation.

For stable sorting, take two input occurrences with equal sort keys at array indices $i<j$. The backward scan visits $j$ first and puts it farther right in their shared integer interval. When $i$ is visited, the endpoint has decreased at least once, so its output position is smaller. Thus their original order survives. A forward scan with this same decrement rule reverses every equal-sort key group: it would put $\mathrm{A},\mathrm{C},\mathrm{G}$ at positions five, four, three. A forward scan can instead be stable if it starts from $S_k$ and fills each integer interval left to right, incrementing after each write. The scan direction and endpoint convention must agree.

Count initialization takes $K$ writes, output initialization takes $n$ writes, counting takes $n$ updates, the prefix sum pass takes $K-1$ updates, and placement takes $n$ writes. Hence time complexity and auxiliary storage are $O(n+K)$ in Big O notation, assuming constant-cost sort key access, integer arithmetic, array indexing and complete sorting record movement. Large records can instead be represented by constant-size references.^[ref]

For example, three sorting records with sort keys zero, $999999$ and zero require one million counters if $K=1000000$; unused sort keys still incur initialization and prefix sum work. The bound is therefore linear in the input length only when $K=O(n)$. Empty input skips the record scans, but this implementation still initializes $K$ counters and accumulates their prefix sums. For nonempty input, counting all $n$ records and filling their ranges gives a concrete termination argument.

Direct sort key indexing is a different operation from a binary comparison. The comparison sorting lower bound concerns comparison sorts; counting sort uses its restricted integer sort keys as array indices. Its small-$K$ speed is particularly useful for categorical numerical labels and for each stable sorting pass of radix sort: the reserved integer intervals turn group sizes into exact destinations.

For $n\ge 0$ sorting records with integer sort keys in $\{0,\dots ,K-1\}$, where $K\ge 1$, let $c_k$ count the input occurrences of sort key $k$. Define two prefix sums:The empty sum gives $S_0=0$. Here $c_k,S_k,P_k$ are counts of sorting records. Exactly $S_k$ sorting records have smaller sort keys, so the output positions for sort key $k$ form the half-open integer intervalIts right endpoint $P_k$ is excluded: the last occupied array index is $P_k-1$ when $c_k>0$.^[ref]

For the seven-sorting record example, the frequency array, inclusive cumulative count array, and start array areThus sort keys zero, one, two and three occupy the integer intervals $[0,2)$, $[2,3)$, $[3,6)$ and $[6,7)$, respectively. In particular, the three sorting records with sort key two belong at array indices three, four and five. The inclusive cumulative count $P_2=6$ counts all six sorting records with sort keys at most two; six is the exclusive endpoint, not the final position of a sorting record with sort key two.

These integer intervals fit together because $P_k=S_{k+1}$ for $k<K-1$, and $P_{K-1}=n$. Their lengths $P_k-S_k=c_k$ add to $n$, so they partition all output positions without overlaps or gaps. An absent sort key contributes an empty integer interval: with frequencies $(0,2,0,1)$, the endpoints are $(0,2,2,3)$ and sort key two has range $[2,2)$. For empty input, every count and endpoint is zero; every reserved integer interval is empty. This makes the same formulas cover missing sort keys and the empty case.

The Cosmic Background Explorer connected two questions: does the cosmic microwave background follow a Planck spectrum, and how does its temperature vary across the sky? John C. Mather and George F. Smoot shared the 2006 Nobel Prize in Physics for these discoveries.^[ref] The FIRAS and COBE DMR comparisons below explain why measuring spectral shape and measuring angular structure require different information.

In 1965, Arno Penzias and Robert Woodrow Wilson reported measurements with a 20-foot horn-reflector antenna at $4.080\mathrm{GHz}$. They manually switched a traveling-wave maser receiver between the horn and a helium-cooled reference. From $6.7\mathrm{K}$ zenith antenna temperature, atmospheric and antenna/ground contributions left an excess $3.5\pm 1.0\mathrm{K}$.^[ref]

Elevation scans estimated the atmospheric term. Taping horn seams produced no measurable change; a nearby transmitter tested ground response. These controls addressed distinct possible contaminations.^[ref]

The comparison does not make all nuisance contributions disappear. A reference can establish a response scale, but atmospheric emission and pickup through unwanted antenna directions are also real incident radiation. In his later account, Robert Woodrow Wilson describes these competing contributions as central to a radio astronomy measurement.^[ref]

The Princeton group proposed primordial thermal radiation as an interpretation and called for more wavelength measurements. Their separate $3\mathrm{cm}$ radiometer had not yet produced results.^[ref] A single-frequency excess therefore does not by itself establish a complete Planck spectrum. Reproducibility across independent instruments, observing directions and frequencies tests the interpretation further.

For a convex set $C$ and $x_0\in C$, a supporting hyperplane has a nonzero normal vector $a$ such that $a^{T}x\le a^T{x}_0$ for all $x\in C$. It touches $C$ while containing the set on one side.

A nonholonomic constraint is a nonintegrable restriction on allowed velocities: it cannot locally be replaced by relations among generalized coordinates and time. In contrast, a velocity equation $r\dot{\varphi }-R\dot{\theta }=0$ with constant radii integrates to a holonomic constraint $r\varphi -R\theta =C$. Rolling without slipping is therefore not automatically nonholonomic: planar rolling on a prescribed curve can be holonomic, whereas rolling with additional directional freedoms can impose nonintegrable velocity constraints. Unwrapped angles may be needed to record the winding of a rolling trajectory.

In Arthur Holly Compton's A Quantum Theory of the Scattering of X-rays by Light Elements (1923), pp. 494–495 and Figure 4 compare X-rays from a molybdenum target with rays scattered by graphite at right angles. Slits select the rays; a calcite crystal analyzes their wavelengths by Bragg's law. The plotted readout is scattered wave intensity against the crystal's glancing angle, distinct from the fixed $90^{\circ }$ photon deflection. The primary-beam comparison uses the same slits, crystal, and tube potential. First-order diffraction lines on both sides establish angular zero. The displaced spectral line supports a longer outgoing wavelength and hence photon energy transfer. The two curves use different vertical scales, so their heights cannot directly compare absolute wave intensities. The paper postpones a fuller apparatus description; it also states on p. 496 that recoil electrons from ordinary X-rays had not been observed. Thus this spectral test infers recoil through the conservation model rather than directly tracking an electron.^[ref]

Inverse Compton scattering transfers energy from an energetic electron to a photon in a Compton scattering event. Transforming into the electron's rest frame blueshifts an approaching photon; scattering there and transforming back can produce a much more energetic outgoing photon.

For a head-on collision with an ultrarelativistic electron of Lorentz factor $\gamma$, the low-recoil limit gives an approximately forward maximum photon energy $E_{\gamma ,\max }\simeq 4{\gamma }^2ϵ$, where $ϵ$ is the incident photon energy. This requires $\gamma ϵ\ll m_e{c}^2$; recoil limits the increase outside that regime.

A laboratory source overlaps a laser beam with an electron beam and selects scattered photons within an angular acceptance. Electron energy spread, beam divergence and collection angle affect the emitted bandwidth, so one photon-energy value does not characterize every collected event. ^[ref]

Choose the incident photon direction as $+x$ and the plane of the collision as the $xy$ plane, with the outgoing photon above the $x$ axis. The electron initially has zero momentum. Conservation of energy and momentum conservation, using the relativistic energy–momentum relation, giveSquaring the electron energy and the momentum magnitude in these two equations yieldsCancel common terms to obtain $m_e{c}^2(E-E^{\prime })=E{E}^{\prime }(1-\cos \theta )$. Division by $E{E}^{\prime }>0$ gives $1/E^{\prime }-1/E=(1-\cos \theta )/(m_e{c}^2)$. Substitution of $\lambda =hc/E$ proves both Compton scattering formulas. Using the nonrelativistic approximation $T_e=|{\mathbf{p}}_e{|}^2/(2m_e)$ here would lose the exact relativistic result.

Let $\varphi$ be the nonnegative electron recoil angle below the incident direction. For $0<\theta <\pi$, the vector components above implyThis assumes exactly the same initially stationary free electron and single outgoing photon. At $\theta =\pi$, the electron moves forward with $\varphi =0$; at $\theta =0$ it stays at rest, so its recoil direction is undefined even though the limiting angle as $\theta \downarrow 0$ is $\pi /2$. At $\theta =90^{\circ }$ in the chosen example, $c{\mathbf{p}}_e=(70,-61.566)\mathrm{keV}$ and $\varphi \simeq 41.33^{\circ }$.

A classical Thomson scattering frequency window for a harmonically bound electron requires ${\omega }_0\ll \omega \ll {\omega }_C$, with ${\omega }_0=\sqrt{k/m}$ and recoil scale ${\omega }_C=m{c}^2/\hslash$. There is no parametrically separated window if ${\omega }_0\gtrsim {\omega }_C$, equivalently $k\gtrsim m^3{c}^4/{\hslash }^2$. This is a scale comparison, not a prediction of a sharp transition between scattering mechanisms.

At that binding scale, an oscillator quantum has energy comparable with the electron rest energy. The nonrelativistic Lorentz oscillator model is then inadequate, so a physical cross-section requires a relativistic quantum description. Compton scattering itself does not switch on discontinuously at $\hslash \omega =m{c}^2$; its recoil corrections become substantial near that scale and tend to the Thomson scattering limit at low photon energy for a free target.

Across a thin steady transition with negligible external impulse, mass conservation and momentum conservation give the displayed relations. If the gas stays isothermal while neutral atomic hydrogen becomes molecular hydrogen, the mean particle mass doubles and $a_2^2=a_1^2/2$ for its isothermal sound speed. Let $J=u_1+a_1^2/u_1$. The downstream subsonic branch has $u_2=[J-\sqrt{J^2-2a_1^2}]/2$, ${\rho }_2/{\rho }_1=u_1/u_2$, and $p_2-p_1={\rho }_1{u}_1(u_1-u_2)>0$.

For $u_1\ll a_1$, the leading behavior is $u_2\simeq u_1/2$, ${\rho }_2\simeq 2{\rho }_1$ and $p_2\simeq p_1$, at unchanged temperature. Radiation of the molecular binding energy prevents that released energy from necessarily heating the gas; it does not, by itself, establish an isothermal thermal closure. If temperature evolution is allowed, an energy balance and a cooling model are required to determine all four changes uniquely.

Let $a>0$, $t>0$ and $z=t+ia/t$. The real and imaginary coordinates obey $XY=a$, a hyperbola. Put $s=t^2-a^2/t^2$, which is strictly increasing. Then $z^2=s+2ia$ traces a horizontal line, and $z^4=s^2-4a^2+4ias$ traces a parabola. Eliminating $s$ gives the displayed quadratic relation between real part and imaginary part. The value of $t$ fixes an orientation and determines which segment is traversed; the full real parabola is obtained as $t$ ranges over $(0,\infty )$.

A phasor is a complex amplitude representing a sinusoid at one specified angular frequency: $a(t)=\mathrm{Re}(\hat{A}{e}^{\mathrm{i}\omega t})$. Its modulus gives the peak amplitude and its complex argument gives the wave phase in this convention. Both may instead use root mean square magnitudes if that normalization is stated.

Excitation of an atomic nucleus by the electric field of a passing charged projectile. For suitable collision energies and separations, the electromagnetic interaction can populate collective states without requiring contact through the nuclear force. Electric quadrupole transition strengths connect excitation yields with collective charge motion; inferring a shape additionally assumes a nuclear model. Contributions from the nuclear force, competing transitions and the collision kinematics must be controlled before identifying a yield with a single quadrupole strength.

A collective oscillation of the quadrupole shape coordinates of an atomic nucleus. Near spherical equilibrium, the ideal harmonic description has five components, corresponding to the five magnetic substates of angular momentum two. Each quantum carries $I^{\pi }=2^{+}$; it is an excitation of the many-nucleon system rather than an extra constituent. Its frequency is $\omega$, and its energy spacing $\hslash \omega$ follows the quantum harmonic oscillator.

One quantum produces a $2^{+}$ level at $\hslash \omega$. Two identical quadrupole quanta can couple to $I=0,2,4$ in a symmetric state, giving a degenerate $0^{+},2^{+},4^{+}$ multiplet at $2\hslash \omega$ in the ideal model. Odd total $I$ would be antisymmetric under their exchange. Thus the first $4^{+}$ to first $2^{+}$ excitation ratio is $2$, rather than the nuclear rotational band value $10/3$. Real nuclear quadrupole vibrations can be anharmonic and their multiplets split: the ratio is a diagnostic of a limiting model, not a universal classification rule.

There is a useful state-counting check. Two identical quanta distributed among five substates span $5(5+1)/2=15$ symmetric states. The proposed multiplet contains $(2\cdot 0+1)+(2\cdot 2+1)+(2\cdot 4+1)=1+5+9=15$. This verifies the angular-state bookkeeping independently of the chosen energy scale.