Seventeen years of hard X-ray observations with the instruments of the INTEGRAL observatory, with a focus on the Milky Way and in particular on the Galactic Centre region, have provided a unique database for exploration of the Galactic population of low-mass X-ray binaries (LMXBs). Our understanding of the diverse energetic phenomena associated with accretion of matter onto neutron stars and black holes has greatly improved. We review the large variety of INTEGRAL based results related to LMXBs. In particular, we discuss the spatial distribution of LMXBs over the Galaxy and their X-ray luminosity function as well as various physical phenomena associated with Atoll and Z sources, bursters, symbiotic X-ray binaries, ultracompact X-ray binaries and persistent black hole LMXBs. We also present an up-to-date catalogue of confirmed LMXBs detected by INTEGRAL, which comprises 166 objects. Last but not least, the long-term monitoring of the Galactic Centre with INTEGRAL has shed light on the activity of Sgr A* in the recent past, confirming previous indications that our supermassive black hole experienced a major accretion episode just ~100 years ago. This exciting topic is covered in this review too.

A systematic search for cosmic gamma-ray bursts (GRBs) and other short hard X-ray events in the archival data from the IBIS/ISGRI telescope of the INTEGRAL observatory over 2003-2018 has been carried out. Seven previously unknown GRBs have been recorded in the telescope field of view; all of them have been localized with an accuracy < 2 arcmin. These events were not revealed by the INTEGRAL burst alert system (IBAS) designed for an automatic GRB search and alert. Four more such localized events missed by IBAS, but known previously, i.e., observed in other experiments, have been found. Eight hundred and eighty six GRBs outside the field of view that arrived at large angles to the IBIS/ISGRI axis have also been recorded. All of them were previously recorded in other experiments, primarily by the anticoincidence shield (ACS) of the SPI gamma-ray spectrometer onboard INTEGRAL, the PICsIT detector of the IBIS gamma-ray telescope, and the KONUS/WIND monitor. An order of magnitude more events without any confirmations in other experiments have been recorded. Both GRBs and solar flares or magnetospheric transient events can be among them. Catalogs with the basic parameters of confirmed and previously unknown GRBs recorded by IBIS/ISGRI have been compiled. The statistical distributions of bursts in various parameters have been constructed and investigated.

Observations of the gravitational-wave (GW) event S190425z registered by the LIGO/Virgo detectors with the Anti-Coincidence Shield (ACS) of the gamma-ray spectrometer SPI aboard the INTEGRAL observatory are presented. With a high probability (>99%) it was associated with a neutron star (NS) merger in a close binary system. This is only the second event of such type in the history of GW observations (after GW170817). A weak gamma-ray burst, GRB190425, consisting of two pulses in ~0.5 and ~5.9 s after the NS merging in S190425z was detected by SPI-ACS. The pulses had a priori reliability of 3.5 and 4.4 sigma as single events and 5.5 sigma as a combined event. Analysis of the SPI-ACS count rate history recorded these days (~125 ks in total) has shown that the rate of appearance of two close pulses with characteristics of GRB190425 by chance does not exceed 6.4 x 10^{-5} s^{-1}. We note that the time profile of GRB190425 has a lot in common with the profile of the GRB170817A burst accompanying the GW170817 event; that both the NS mergers were the closest (<150 Mpc) of all the events registered by the LIGO/Virgo detectors; and that there were no confident excesses of gamma-ray emission over the background detected in any of >30 black hole merger events recorded to the moment by these detectors. No hard X-ray flares were detected in the field of view of the SPI and IBIS-ISGRI gamma-ray telescopes aboard INTEGRAL. This, as well as the lack of detection of gamma-ray emission from GRB190425 by the GBM monitor of the Fermi observatory assuming its occultation by the Earth, can significantly reduce the localization area for the source of this GW event. The estimates of the parameters E_{iso} and E_{p} for GRB190425 are obtained and compared with the similar parameters for GRB170817A.

The excess of the rate of type I X-ray bursts over that expected when the matter fallen between bursts completely burns out in a thermonuclear explosion is explained in terms of the model of a spreading layer of matter coming from the accretion disk over the neutron star surface. Such excess is observed in bursters with a high persistent luminosity, $4\times 10^{36}\ \mbox{erg s}^{-1}\leq L_{X}\leq 2\times 10^{37}\ \mbox{erg s}^{-1}$. In this model the accreting matter settles to the stellar surface mainly in two high-latitude ring zones. Despite the subsequent spreading of matter over the entire star, its surface density in these zones turns out to be higher than the average one by 2-3 orders of magnitude, which determines the predominant ignition probability. The multiple events whereby the flame after the thermonuclear explosion in one ring zone (initial burst) propagates through less dense matter to another zone and initiates a second explosion in it (recurrent burst) make a certain contribution to the observed excess of the burst rate. However, the localized explosions of matter in these zones, after which the burning in the zone rapidly dies out without affecting other zones, make a noticeably larger contribution to the excess of the burst rate over the expected one.

We report the detection during the JEM-X/INTEGRAL observations of several X-ray bursters of series of close type I X-ray bursts consisting of two or three events with a recurrence time much shorter than the characteristic (at the observed mean accretion rate) time of matter accumulation needed for a thermonuclear explosion to be initiated on the neutron star surface. We show that such series of bursts are naturally explained in the model of a spreading layer of accreting matter over the neutron star surface in the case of a sufficiently high ($\dot{M}\geq 1\times 10^{-9}\ M_{\odot}\ \mbox{yr}^{-1}$) accretion rate (corresponding to a mean luminosity $L_{\rm tot}\geq 1\times 10^{37} \mbox{erg s}^{-1}$). The existence of triple bursts requires some refinement of the model - the importance of a central ring zone is shown. In the standard model of a spreading layer no infall of matter in this zone is believed to occur.

To find X-ray bursts from sources within the field of view of the IBIS/INTEGRAL telescope, we have analysed all the archival data of the telescope available at the time of writing the paper (the observations from January 2003 to April 2009). We have detected 834 hard (15-25 keV) X-ray bursts, 239 of which were simultaneously recorded by the JEM-X/INTEGRAL telescope in the standard X-ray energy range. More than 70% of all bursts (587 events) have been recorded from the well-known X-ray burster GX 354-0. We have found upper limits on the distances to their sources by assuming that the Eddington luminosity limit was reached at the brightness maximum of the brightest bursts.

We report results of the INTEGRAL Target of Opportunity observations of the transient X-ray burster XTE J1810-189. The observations were performed on April 3--6, 2008, soon after the discovery of the source and near the peak of its outburst. That time the source had a flux of about 50 mCrab and exhibited a hard Comptonized X-ray spectrum extending well above 100 keV. Being approximated by a power law with an exponetial cut-off in the broad 3--100 keV energy band it gave the average photon index $\Gamma\simeq 1.6$ and $kT_{cutoff}\simeq 67$ keV. We found only slight indications for changes in the index during the observation ($\Gamma$ first steady decreased from $\sim 2.0$ to $\sim1.3$ and then increased back to $\sim 2.0$). However the $N_{\rmn H}$ value measured by absorption in the low energy part of the spectrum changed drastically and very irregularly (from $\sim 4\times 10^{22}$ till $\sim 100\times 10^{22}$ cm$^{-2}$). There were 10 type I X-ray bursts detected from the source during these TOO observations. Assuming that the Eddington luminosity was reached during the burst with the highest peak flux we get an upper estimate for a distance to the source $D=6.4\pm0.6$ kpc. From the X-ray burst parameters we conclude that this LMXB harboures an evolved star.

During the analysis of the INTEGRAL observatory archival data we found a powerful X-ray burst, registered by JEM-X and IBIS/ISGRI telescopes on April 16, 2005 from a weak and poorly known source AX J1754.2-2754. Analysis of the burst profiles and spectrum shows, that it was a type I burst, which result from thermonuclear explosion on the surface of nutron star. It means that we can consider AX J1754.2-2754 as an X-ray burster. Certain features of burst profile at its initial stage witness of a radiation presure driven strong expansion and a corresponding cooling of the nutron star photosphere. Assuming, that the luminosity of the source at this phase was close to the Eddington limit, we estimated the distance to the burst source d=6.6+/-0.3 kpc (for hidrogen atmosphere of the neutron star) and d=9.2+/-0.4 kpc (for helium atmosphere).

Analyzing archival data of the INTEGRAL observatory we discovered and localized a cosmic gamma-ray burst detected within fields of view of the IBIS/ISGRI and SPI telescopes on April 28, 2006. The burst has not been revealed by the INTEGRAL Burst Alert System (IBAS), so the information on its coordinates has not been distributed in time and search for the afterglow has not been carried out. The burst has been also detected by the KONUS/WIND and RHESSI spacecrafts. Its fluence was 2.3x10^{-6} erg/cm^2 in the 20-200 keV energy band, the maximum flux was 3.6x10^{-7} erg/cm^2/s (3.9 phot/cm^2/s). The burst had a complicated multi-peak profile and was outstanding of the typical bursts by increasing its spectral hardness with time. The emission spectrum near the flux maximum was characterized by the photon index alpha ~ -1.5 and the peak energy Ep ~ 95 keV. The burst lasted ~12 s, then we observed its afterglow at energies 15-45 keV decaying according to a power law with the index ~ -4.5. The spectral hardness decreased substantially during the afterglow.

All the observations performed with the IBIS telescope aboard the INTEGRAL observatory during the first 2.5 years of its in-orbit operation have been analyzed to find X-ray bursts. There were 1788 statistically confident events with a duration from 5 to 500 s revealed in time records of the 15-25 keV count rate of the IBIS/ISGRI detector, 319 of them were localized and, with one exception, identified with persistent X-ray sources. The known bursters were responsible for 215 of the localized events. One burst was detected from AXJ1754.2-2754, the source previously unknown as a burster, and another burst - from a new source. There was duality in determining its position - its name could be either IGR J17364-2711 or IGR J17380-3749. Curiously enough, the 138 bursts were detected from one X-ray burster - GX 354-0.

All of the observations performed with the IBIS telescope onboard the INTEGRAL observatory during the first one and a half years of its in-orbit operation (from February 10, 2003, through July 2, 2004) have been analyzed to find X-ray bursts. IBIS/ISGRI detector lightcurves total count rate in the energy range 15-25 keV revealed 1077 bursts of durations from ~5 to ~500 s detected with a high statistical significance (only one event over the entire period of observations could be detected by a chance with a probability of 20%). Apart from the events associated with cosmic gamma-ray bursts (detected in the field of view or passed through the IBIS shield), solar flares, and activity of the soft gamma repeater SGR1806-20, we were able to localize 105 bursts and, with one exception, to identify them with previously known persistent or transient X-ray sources (96 were identified with known X-ray bursters). In one case, the burst source was a new burster in a low state. We named it IGR J17364-2711. Basic parameters of the localized bursts and their identifications are presented in the catalog of bursts. Curiously enough, 61 bursts were detected from one X-ray burster - GX 354-0. The statistical distributions of bursts in duration, maximum flux, and recurrence time have been analyzed for this source. Some of the bursts observed with the IBIS/ISGRI telescope were also detected by the JEM-X telescope onboard the INTEGRAL observatory in the standard X-ray energy range 3-20 keV.

We present the results of our study of the emission from the transient burster MX 0836-42 using its observations by the INTEGRAL and RXTE X-ray and gamma-ray observatories in the period 2003-2004. The source's broadband X-ray spectrum in the energy range 3-120 keV has been obtained and investigated for the first time. We have detected 39 X-ray bursts from this source. Their analysis shows that the maximum 3-20 keV flux varies significantly from burst to burst, F ~ (0.5-1.5)x10^{-8} erg cm^{-2} s^{-1}. Using the flux at the maximum of the brightest detected burst, we determined an upper limit for the distance to the source, D ~ 8 kpc.